Method and device for adjusting distance from steel, electronic equipment and storage medium

Abstract

This invention relates to the field of steel rolling technology, and more particularly to a method, device, electronic equipment, and storage medium for adjusting the distance between the billet and the steel. The method first obtains the first throwing time and the first biting time; then, based on the speed of the rolling mill production line, the first throwing time, and the first biting time, a first cutting length is determined; next, based on the on-site process parameters of the rolling mill production line, a first current dataset, and a second current dataset, a cutting length analysis is performed to obtain the first cutting length and the second cutting length; finally, based on the first cutting length, the first cutting length, and the second cutting length, the tail-end cutting length of the first billet and the head-end cutting length of the second billet are determined. This invention uses principal component analysis to remove redundancy and a defect analysis model to accurately identify defects, achieving precise positioning of defect areas at the billet ends, ensuring continuous and stable operation of the rolling mill production line, and improving overall production capacity.

Description

Technical Field

[0001] This invention relates to the field of steel rolling technology, and in particular to a method, device, electronic device, and storage medium for adjusting the distance to the steel. Background Technology

[0002] In the continuous rolling process of a rolling mill production line, steel billets must enter the mill continuously for rolling to ensure production efficiency and capacity. The cutting quality at the joint of the steel billet directly determines the precision, surface quality, and mechanical properties of the subsequently rolled products. Improper cutting can easily lead to rolling defects such as residual cracks and flat tails at the ends of the steel billet, or waste of steel due to over-cutting. It may also affect the normal operation of the rolling mill equipment, increasing production safety hazards and equipment maintenance costs.

[0003] Currently, the cutting method at the joint of steel billets in rolling mill production lines mostly adopts a fixed parameter cutting mode. When using the fixed parameter cutting mode, a fixed cutting length is usually preset, without considering factors such as the steel billet joint sequence, equipment reset status, real-time rolling conditions, and the actual defect distribution at the end of the steel billet. This leads to a mismatch between the cutting length and the actual needs: on the one hand, if the fixed cutting length is insufficient, the defects at the joint of the steel billet cannot be completely removed, and the residual defects will further expand in the subsequent rolling process; on the other hand, if the fixed cutting length is too large, it will cause a large amount of steel waste and increase production costs.

[0004] Furthermore, existing cutting methods lack precise control over the timing of billet connection, failing to fully consider the time interval between the first billet being thrown and the second billet being bitten, as well as the reset time of cutting-related equipment such as flying shears and loopers. This can easily lead to cutting operations being performed before the equipment has fully reset, causing minor issues like equipment jamming and wear, or even serious safety accidents, affecting the continuity and stability of the production line. Simultaneously, existing technologies lack efficient defect identification and analysis methods, making it impossible to accurately locate defective areas at the billet ends, hindering fine-tuning of the cutting length, and exhibiting low levels of intelligence. They cannot dynamically adapt cutting strategies based on real-time production data, failing to meet the refined, efficient, and intelligent production requirements of modern rolling mill production lines.

[0005] Therefore, in response to the technical problems existing in the current billet cutting technology of rolling mills, such as low cutting accuracy, inaccurate defect identification, serious steel waste, low production efficiency, large equipment wear and tear and low level of intelligence, there is an urgent need for a billet cutting method that can accurately control cutting parameters, accurately identify defects, adapt to different working conditions, and balance quality and efficiency, so as to solve the shortcomings of the existing technology and promote the technological upgrading of rolling mill production lines. Summary of the Invention

[0006] The present invention provides a method, device, electronic device and storage medium for adjusting the distance between the steel and the steel, which solves the problem of difficulty in balancing equipment safety and production efficiency in the prior art.

[0007] In a first aspect, embodiments of the present invention provide a method for adjusting the distance to the steel beam, comprising: The first steel throwing time and the first steel biting time are obtained, wherein the first steel throwing time and the first steel biting time correspond to the first steel billet and the second steel billet, respectively, and the second steel billet enters the rolling mill production line after the first steel billet; Based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, the first cutting length is determined, wherein the first cutting length is the sum of the tail cutting length of the first steel billet and the head cutting length of the second steel billet. Based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset, the cutting length is analyzed to obtain the first cutting length and the second cutting length. The first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. Based on the first cutting length and the second cutting length, the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet are determined.

[0008] In one possible implementation, determining the first cutting length based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time includes: The difference between the first steel biting time and the first steel throwing time is taken as the steel passing interval time; The difference between the steel-passing interval time and the maximum equipment reset time is taken as the first interval time difference, wherein the maximum equipment reset time is the time determined based on the flying shear reset time and the maximum value among multiple looper reset times; Multiply the first time interval difference by the speed of the rolling mill production line to obtain the first cutting length.

[0009] In one possible implementation, the step of performing cutting length analysis based on the on-site process parameters of the rolling mill production line, a first current dataset, and a second current dataset to obtain a first cutting length and a second cutting length includes: For each current dataset in the first current dataset and the second current dataset, perform the following steps respectively: From the current dataset, extract the current parameter array of the rolling mill sequentially; The extracted current parameter array and the on-site process parameters are used to construct the first vector to be analyzed. The second vector to be analyzed is obtained by performing principal component analysis on the first vector to be analyzed using a dimensionality reduction matrix. The dimensionality reduction matrix is ​​obtained by performing principal component analysis on multiple historical vectors to be analyzed. The second vector to be analyzed is input into the defect analysis model to obtain the rolling defect probability; Add the obtained rolling defect probability to the defect probability queue; If the traversal of the current dataset is completed, the first cutting length or the second cutting length is determined according to the probability threshold and the defect probability queue. Otherwise, proceed to the step of sequentially extracting the current parameter array of the rolling mill from the current dataset.

[0010] In one possible implementation, the dimensionality reduction matrix is ​​obtained by performing principal component analysis on multiple historical vectors to be analyzed, including: Obtain the multiple historical vectors to be analyzed; The multiple historical vectors to be analyzed are standardized according to their dimensions to obtain multiple standard vectors; Perform cocorrelation analysis on the multiple standard vectors to obtain the cocorrelation matrix; Extract multiple eigenvalues ​​from the cocorrelation matrix; Multiple eigenvectors of the cocorrelation matrix are extracted based on the multiple eigenvalues, wherein each eigenvalue corresponds to one eigenvector; Multiple target values ​​are selected from the plurality of feature values ​​in descending order, wherein the ratio of the sum of the multiple target values ​​to the sum of the multiple feature values ​​is greater than the principal component threshold. Use the feature vector corresponding to the target value as the target vector; Multiple target vectors are used to construct the dimensionality reduction matrix.

[0011] In one possible implementation, the standardization of the plurality of historical vectors to be analyzed according to dimensions to obtain a plurality of standard vectors includes: Perform the following steps for each dimension: Based on the dimensions, dimension data is extracted from the multiple historical vectors to be analyzed and constructed into a dimension array; Calculate the mean and standard deviation of multiple data points in the dimensional array; Based on the first formula, the mean of the multiple data points, and the standard deviation of the multiple data points, the dimensional array is standardized to obtain a standard array, wherein the first formula is:

[0012] In the formula, For the first in the standard array One data point, For the 1st dimension in the array One data point, The mean of multiple data points. The standard deviation of multiple data points; Based on the correspondence with the dimension array, the data in the standard array is added to the standard vector.

[0013] In one possible implementation, the defect analysis model is constructed based on multiple historical analysis vectors, including: Multiple historical analysis vectors are obtained, where each historical analysis vector corresponds to an identifier representing the existence of rolling defects. The historical analysis vectors are obtained by performing principal component analysis on the historical vectors to be analyzed using the dimensionality reduction matrix. Substitute the aforementioned historical analysis vectors into the basic model to obtain multiple defect probability estimates; The fitting loss is determined based on the multiple defect probability estimates and the identifiers of the multiple historical analysis vectors; If the fitting loss is greater than the loss threshold, the parameters of the basic model are adjusted according to the fitting loss, and the process jumps to the step of substituting the multiple historical analysis vectors into the basic model to obtain multiple probability values. Otherwise, the basic model shall be used as the defect analysis model; The basic model is as follows:

[0014] In the formula, This is a probability estimate. It is a natural constant. The intercept is... For the first One regression coefficient, To analyze vectors, This is to analyze the total number of elements in the vector.

[0015] In one possible implementation, determining the tail end cutting length of the first billet and the head end cutting length of the second billet based on the first cutting length and the second cutting length includes: If the sum of the first cutting length and the second cutting length is greater than the sum of the first cutting lengths, then the tail end of the first steel billet is cut according to the first cutting length, and the head end of the second steel billet is cut according to the second cutting length. Otherwise, based on the first cutting length and the readjusted tail-end cutting length of the first billet and the head-end cutting length of the second billet, wherein the adjusted tail-end cutting length of the first billet is greater than the first cutting length, and the adjusted head-end cutting length of the second billet is greater than the second cutting length.

[0016] Secondly, embodiments of the present invention provide a steel-following distance adjustment device for implementing the steel-following distance adjustment method as described in the first aspect or any possible implementation thereof, the steel-following distance adjustment device comprising: The steel passing time acquisition module is used to acquire the first steel throwing time and the first steel biting time, wherein the first steel throwing time and the first steel biting time correspond to the first steel billet and the second steel billet, respectively, and the second steel billet enters the rolling mill production line after the first steel billet; The minimum cutting length acquisition module is used to determine the first cutting length sum based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, wherein the first cutting length sum is the sum of the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet; The defect length determination module is used to perform cutting length analysis based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset to obtain the first cutting length and the second cutting length. The first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. as well as, The cutting length determination module is used to determine the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet based on the first cutting length and the second cutting length.

[0017] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the steps of the method as described in the first aspect or any possible implementation of the first aspect.

[0018] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any possible implementation thereof.

[0019] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: This invention discloses a method for adjusting the distance between the billet and the steel billet. First, it acquires a first billet throwing time and a first steel biting time, where the first billet throwing time and the first steel biting time correspond to a first steel billet and a second steel billet, respectively. The second steel billet enters the rolling mill production line after the first steel billet. Then, based on the speed of the rolling mill production line, the first billet throwing time, and the first steel biting time, it determines a first cutting length sum, where the first cutting length sum is the sum of the tail cutting length of the first steel billet and the head cutting length of the second steel billet. Next, based on the on-site process parameters of the rolling mill production line, a first current dataset, and a second... The current dataset is used to perform cutting length analysis to obtain a first cutting length and a second cutting length, wherein the first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. Finally, based on the first cutting length and the second cutting length, the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet are determined.

[0020] This invention combines on-site process parameters and current datasets, and uses principal component analysis to remove redundancy and a defect analysis model to accurately identify defects, thereby achieving precise positioning of the defect area at the end of the billet and determining a reasonable cutting length. Through dual judgment logic, it takes into account both defect removal and length benchmark, effectively avoiding the problems of insufficient cutting (residual defects) or over-cutting (poor end flatness) caused by inaccurate defect identification and length calculation deviation in traditional cutting. This significantly improves the end quality of the billet after cutting, provides qualified billets for subsequent rolling processes, and reduces the product scrap rate caused by cutting quality problems.

[0021] This invention improves production efficiency and ensures rolling continuity. The method of this invention clearly defines the billet connection sequence, ensuring that the second billet precisely connects with the first billet as they enter the production line, avoiding production interruptions. Scientific calculations determine the cutting length, providing a clear basis for the cutting operation and reducing waiting and adjustment time during the cutting process. Automated data analysis and defect identification processes replace traditional manual defect judgment, significantly shortening the time spent on cutting length analysis and reducing the subjective error of manual judgment. Each step forms a closed-loop connection, effectively improving the automation level and operational efficiency of the cutting operation, ensuring the continuous and stable operation of the rolling mill production line, and increasing overall production capacity. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of the steel-following distance adjustment method provided in the embodiments of the present invention; Figure 2 This is a schematic diagram of a rolling mill production line provided in an embodiment of the present invention; Figure 3 This is a functional block diagram of the steel-following distance adjustment device provided in the embodiments of the present invention; Figure 4 This is a functional block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] In the following description, specific details such as particular system structures and techniques are set forth for illustrative purposes and not for limitation, so as to provide a thorough understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0025] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0026] The embodiments of the present invention will be described in detail below. This example is implemented based on the technical solution of the present invention, and provides detailed implementation methods and specific operation processes. However, the protection scope of the present invention is not limited to the following embodiments.

[0027] Figure 1A flowchart of the steel-following distance adjustment method provided for embodiments of the present invention.

[0028] like Figure 1 As shown, a flowchart illustrating the implementation of the steel-following distance adjustment method provided by an embodiment of the present invention is presented, and is described in detail below: In step 101, the first steel throwing time and the first steel biting time are obtained, wherein the first steel throwing time and the first steel biting time correspond to the first steel billet and the second steel billet, respectively, and the second steel billet enters the rolling mill production line after the first steel billet.

[0029] In step 102, a first cutting length sum is determined based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, wherein the first cutting length sum is the sum of the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet.

[0030] In some embodiments, determining the first cutting length based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time includes: The difference between the first steel biting time and the first steel throwing time is taken as the steel passing interval time; The difference between the steel-passing interval time and the maximum equipment reset time is taken as the first interval time difference, wherein the maximum equipment reset time is the time determined based on the flying shear reset time and the maximum value among multiple looper reset times; Multiply the first time interval difference by the speed of the rolling mill production line to obtain the first cutting length.

[0031] For example, such as Figure 2 As shown, the rolling mill production line is a continuous rolling production line consisting of multiple rolling mills 201. Typically, a continuous rolling production line is divided into multiple rolling mill groups. For example, a rolling mill production line may have a total of 18 rolling mills 201, with the first 9 mills 201 forming one group and the last 9 mills 201 forming another. Before the steel billet 203 is transferred from the first group to the second group, a portion of its beginning and end is cut off by a flying shear 202. This serves two purposes: firstly, to cut off defective parts and prevent them from amplifying during subsequent rolling processes; secondly, due to the extension effect of the beginning and end, the distance between the two connecting steel billets 203 will be reduced to some extent. Reasonably allocating the cutting length at the beginning and end can increase the distance between the two steel billets 203, providing reaction time for equipment (such as looper reset and flying shear reset) and preventing equipment safety accidents.

[0032] However, excessive cutting lengths lead to material waste and decreased production efficiency. A reasonable cutting length should be determined in conjunction with the steel-passing interval; that is, the cutting length should be minimized as much as possible while ensuring equipment safety and that defective parts are completely cut off.

[0033] As can be seen from the above, the cutting accuracy at the joint directly affects product quality and production efficiency. To clarify the specific implementation process of the cutting operation, each step is explained in detail below: This invention uses a real-time monitoring system on the rolling mill production line to accurately collect the first billet ejection time and the first billet bite time. In some embodiments, the ejection time and bite time are identified by changes in the rolling mill current. The first ejection time specifically refers to the moment when the first billet completely leaves the rolling mill roll system, completes rolling, and leaves the rolling mill working area. The first bite time specifically refers to the moment when the second billet is clamped by the rolling mill roll system and begins to enter the rolling state. The first ejection time and the first bite time correspond one-to-one with the first billet and the second billet, respectively. The second billet is continuously connected, entering the rolling mill production line just as the tail end of the first billet is about to leave the rolling mill working area, ensuring the continuity of the rolling process and avoiding production interruptions.

[0034] Based on the stable rolling speed of the rolling mill production line, and combined with the first steel throwing time and the first steel biting time obtained in the aforementioned steps, the first cutting length is determined through specific calculation logic. The first cutting length is a key comprehensive parameter, specifically defined as the sum of the length to be cut from the tail end of the first steel billet and the length to be cut from the head end of the second steel billet. Its core function is to provide a basic length benchmark for subsequent precise cutting, ensuring that defects at the junction of the two steel billets are completely removed, while avoiding steel waste caused by over-cutting.

[0035] The first cutting length is determined based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, specifically including the following steps: Calculate the steel passing interval time: Calculate the difference between the first steel biting time and the first steel throwing time. The result is the steel passing interval time between the two steel billets. This time reflects the time gap between the first steel billet leaving and the second steel billet entering the rolling mill working area. It is the core basis for subsequent calculations.

[0036] Calculate the first interval time difference: Subtract the maximum equipment reset time from the steel passing interval time to obtain the first interval time difference. The maximum equipment reset time is a key parameter ensuring the normal operation of the cutting equipment. Its specific value is determined based on the flying shear reset time and the maximum value among the reset times of multiple loopers in the production line. The flying shear reset time refers to the time required for the flying shear to return to its initial cutting position after completing the previous cutting action, while the looper reset time refers to the time required for each looper to adjust to the preset tension state and ensure rolling stability. Taking the maximum of the two ensures that the cutting equipment has sufficient reset time, avoiding cutting deviations caused by incomplete equipment reset.

[0037] Calculate the first cutting length sum: Multiply the first time interval difference by the stable rolling speed of the rolling mill production line, and the product is the first cutting length sum. Since the rolling mill production line has a constant speed during stable operation, the distance corresponding to the time difference is the theoretical redundancy length at the junction of the two steel billets, which is the total length to be cut.

[0038] For example, assuming the stable rolling speed of the rolling mill production line is 3 m / s, the first steel throwing time is 100.0 s, and the first steel biting time is 101.0 s, the steel passing interval time can be calculated to be 5.0 s. If the flying shear reset time is 1.2 s, and the reset times of the three loopers in the production line are 0.8 s, 1.0 s, and 1.1 s respectively, then the maximum equipment reset time is taken as 1.2 s. Further calculation shows that the first interval time difference is 1.2 s - 1 s = 0.2 s, and the final first cutting length is 0.2 s × 8 m / s = 1.6 m, that is, the total cutting length of the tail end of the first steel billet and the head end of the second steel billet needs to reach 1.6 m.

[0039] In step 103, the cutting length is analyzed based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset to obtain the first cutting length and the second cutting length. The first cutting length and the second cutting length correspond to the tail end of the first billet and the head end of the second billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first billet and the head end of the second billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals.

[0040] In some embodiments, the step of performing cutting length analysis based on the on-site process parameters of the rolling mill production line, a first current dataset, and a second current dataset to obtain a first cutting length and a second cutting length includes: For each current dataset in the first current dataset and the second current dataset, perform the following steps respectively: From the current dataset, extract the current parameter array of the rolling mill sequentially; The extracted current parameter array and the on-site process parameters are used to construct the first vector to be analyzed. The second vector to be analyzed is obtained by performing principal component analysis on the first vector to be analyzed using a dimensionality reduction matrix. The dimensionality reduction matrix is ​​obtained by performing principal component analysis on multiple historical vectors to be analyzed. The second vector to be analyzed is input into the defect analysis model to obtain the rolling defect probability; Add the obtained rolling defect probability to the defect probability queue; If the traversal of the current dataset is completed, the first cutting length or the second cutting length is determined according to the probability threshold and the defect probability queue. Otherwise, proceed to the step of sequentially extracting the current parameter array of the rolling mill from the current dataset.

[0041] In some implementations, the dimensionality reduction matrix is ​​obtained by performing principal component analysis on multiple historical vectors to be analyzed, including: Obtain the multiple historical vectors to be analyzed; The multiple historical vectors to be analyzed are standardized according to their dimensions to obtain multiple standard vectors; Perform cocorrelation analysis on the multiple standard vectors to obtain the cocorrelation matrix; Extract multiple eigenvalues ​​from the cocorrelation matrix; Multiple eigenvectors of the cocorrelation matrix are extracted based on the multiple eigenvalues, wherein each eigenvalue corresponds to one eigenvector; Multiple target values ​​are selected from the plurality of feature values ​​in descending order, wherein the ratio of the sum of the multiple target values ​​to the sum of the multiple feature values ​​is greater than the principal component threshold. Use the feature vector corresponding to the target value as the target vector; Multiple target vectors are used to construct the dimensionality reduction matrix.

[0042] In some implementations, the standardization of the plurality of historical vectors to be analyzed according to dimensions to obtain a plurality of standard vectors includes: Perform the following steps for each dimension: Based on the dimensions, dimension data is extracted from the multiple historical vectors to be analyzed and constructed into a dimension array; Calculate the mean and standard deviation of multiple data points in the dimensional array; Based on the first formula, the mean of the multiple data points, and the standard deviation of the multiple data points, the dimensional array is standardized to obtain a standard array, wherein the first formula is:

[0043] In the formula, For the first in the standard array One data point, For the 1st dimension in the array One data point, The mean of multiple data points. The standard deviation of multiple data points; Based on the correspondence with the dimension array, the data in the standard array is added to the standard vector.

[0044] In some implementations, the defect analysis model is constructed based on multiple historical analysis vectors, including: Multiple historical analysis vectors are obtained, where each historical analysis vector corresponds to an identifier representing the existence of rolling defects. The historical analysis vectors are obtained by performing principal component analysis on the historical vectors to be analyzed using the dimensionality reduction matrix. Substitute the aforementioned historical analysis vectors into the basic model to obtain multiple defect probability estimates; The fitting loss is determined based on the multiple defect probability estimates and the identifiers of the multiple historical analysis vectors; If the fitting loss is greater than the loss threshold, the parameters of the basic model are adjusted according to the fitting loss, and the process jumps to the step of substituting the multiple historical analysis vectors into the basic model to obtain multiple probability values. Otherwise, the basic model shall be used as the defect analysis model; The basic model is as follows:

[0045] In the formula, This is a probability estimate. It is a natural constant. The intercept is... For the first One regression coefficient, To analyze vectors, This is to analyze the total number of elements in the vector.

[0046] For example, by combining the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset, a refined cutting length analysis is conducted to ultimately obtain the first cutting length and the second cutting length. The first cutting length corresponds to the tail-end cutting dimension of the first steel billet, and the second cutting length corresponds to the head-end cutting dimension of the second steel billet. The first current dataset is the set of current parameters collected by the rolling mill during the rolling process at the tail end of the first steel billet, and the second current dataset is the set of current parameters collected by the rolling mill during the rolling process at the head end of the second steel billet. Each current dataset contains an array of rolling mill current parameters continuously collected at preset fixed time intervals (e.g., 10ms). Changes in the current parameters can directly reflect the rolling quality of the billet end, such as the presence of defects like cracks or inclusions, providing data support for determining the cutting length.

[0047] Based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset, a cutting length analysis is performed to obtain the first cutting length and the second cutting length. This specifically includes the following steps, which are performed independently for each current dataset in the first and second current datasets: Extracting Current Parameter Array: From the currently processed current dataset, extract the current parameter arrays of the rolling mill sequentially according to time sequence. Constructing the First Analysis Vector: Merge the extracted current parameter arrays with the on-site process parameters to construct the first analysis vector. The on-site process parameters include key parameters such as roll speed, rolling temperature, billet material, and reduction. Combining these with the current parameters comprehensively reflects the rolling state of the billet ends, improving the accuracy of the analysis.

[0048] Principal Component Analysis (PCA) Processing: PCA is performed on the first vector to be analyzed using a pre-defined dimensionality reduction matrix, transforming the high-dimensional vector into a low-dimensional second vector. The dimensionality reduction matrix is ​​pre-constructed after analyzing a large number of historical vectors using PCA. Its core function is to remove redundant information from the data, retain the key features that best reflect the defects in the steel billet, reduce the computational load of subsequent analysis, and improve the accuracy of the analysis.

[0049] Obtaining the rolling defect probability: The processed second vector to be analyzed is input into the pre-trained defect analysis model. Through the model's calculation and analysis, the corresponding rolling defect probability is output. This probability is used to characterize the likelihood of rolling defects existing in the end region of the steel billet currently being analyzed.

[0050] Construct a defect probability queue: Add the obtained rolling defect probabilities to the defect probability queue in the order of data acquisition time to form a continuous defect probability sequence, which facilitates the overall judgment of the defect area at the end of the billet.

[0051] Determine the traversal status and cut length: If the traversal of the current current dataset has been completed, that is, all current parameters collected at the end of the billet have been analyzed, then the first cut length or the second cut length is determined according to the preset probability threshold and the defect probability queue. When the defect probability at a certain position in the defect probability queue exceeds the probability threshold, it indicates that there is a defect in the billet area corresponding to that position, and the area needs to be included in the cutting range. Finally, the cut length is determined. If the traversal of the current current dataset has not been completed, then the process jumps to the step of extracting the current parameter array of the rolling mill from the current dataset in sequence and continues the analysis until the traversal of all data is completed.

[0052] Methods for constructing dimensionality reduction matrices: The dimensionality reduction matrix is ​​obtained by performing principal component analysis on multiple historical vectors to be analyzed. The specific steps include: Acquire historical data: Collect multiple historical vectors to be analyzed accumulated during the operation of the rolling mill production line. These historical vectors need to cover different production conditions, billet materials, rolling parameters and other scenarios to ensure the diversity and representativeness of the data, and provide a sufficient sample basis for the construction of the dimensionality reduction matrix.

[0053] Vector standardization: The multiple historical vectors to be analyzed are standardized according to their different dimensions, resulting in multiple standard vectors. The core purpose of standardization is to eliminate dimensional differences between data from different dimensions, preventing the accuracy of the analysis results from being affected by the excessively large numerical range of a single dimension.

[0054] Calculating the cocorrelation matrix: Perform cocorrelation analysis on multiple standardized vectors to calculate the cocorrelation matrix. The cocorrelation matrix reflects the degree of linear correlation between data in different dimensions. The larger the element value in the matrix, the stronger the correlation between the corresponding two dimensions, and vice versa.

[0055] Feature extraction: Multiple feature values ​​are extracted from the cocorrelation matrix using the eigenvalue decomposition method. The magnitude of the feature value reflects the total amount of original data information contained in the corresponding feature vector. The larger the feature value, the better the corresponding feature vector reflects the core features of the original data.

[0056] Feature vector extraction: Based on the extracted multiple feature values, multiple feature vectors corresponding to the cocorrelation matrix are extracted. Each feature value corresponds to a unique feature vector. These feature vectors are orthogonal to each other and constitute the feature space of the original data.

[0057] Selecting target feature values: From the extracted feature values, select multiple target values ​​in descending order. The selection criterion is that the ratio of the sum of the multiple target values ​​to the sum of all feature values ​​(i.e., the cumulative contribution rate) is greater than the preset principal component threshold (usually set to 85%~95%), ensuring that the selected target feature values ​​can retain most of the information of the original data.

[0058] Constructing the dimensionality reduction matrix: Using the feature vectors corresponding to the selected target values ​​as target vectors, all target vectors are arranged in order to construct the dimensionality reduction matrix. This matrix can effectively reduce the dimensionality of high-dimensional vectors to be analyzed while preserving key information from the original data to the maximum extent.

[0059] The specific implementation of vector normalization: The multiple historical vectors to be analyzed are standardized according to their dimensions to obtain multiple standard vectors. This process includes the following steps, which must be performed independently for each dimension: Extract dimensional data: Based on the dimension being processed, extract all data corresponding to that dimension from multiple historical vectors to be analyzed, organize them into a dimension array, and ensure that the array contains all historical data under that dimension without omissions or errors.

[0060] Calculate statistical parameters: Based on the constructed dimensional array, calculate the mean and standard deviation of all data in the array. The mean is the arithmetic mean of all data in the array, reflecting the central tendency of the data in this dimension; the standard deviation is the square root of the average of the sum of the squares of the differences between all data in the array and the mean, reflecting the dispersion of the data in this dimension.

[0061] Standardization calculation: Based on the first formula, the mean of multiple data points, and the standard deviation of the multiple data points, each data point in the dimension array is standardized to obtain a standard array. The first formula is: Constructing standard vectors: Based on the order of the data in the dimension array, add the data from the standard array obtained after standardization to the corresponding positions in the standard vector to complete the standardization process for that dimension; repeat the above steps until all dimensions are standardized, and finally obtain multiple standard vectors.

[0062] Methods for constructing defect analysis models: The defect analysis model is constructed based on multiple historical analysis vectors, and specifically includes the following steps: Obtain historical analysis vectors: Collect multiple historical analysis vectors, each corresponding to a specific identifier indicating the presence or absence of rolling defects (e.g., "1" indicates the presence of defects, and "0" indicates the absence of defects). These historical analysis vectors are low-dimensional vectors obtained by performing principal component analysis on the historical vectors to be analyzed using the dimensionality reduction matrix. Compared to the historical vectors to be analyzed, these vectors better highlight the key features of billet defects.

[0063] Model prediction calculation: Substitute the multiple historical analysis vectors into the preset basic model, and obtain the defect probability estimate corresponding to each historical analysis vector through model calculation. The estimate is used to characterize the possibility of defects in the billet area corresponding to the historical analysis vector.

[0064] Calculate the fitting loss: Based on the multiple defect probability estimates and the corresponding historical analysis vectors, calculate the model's fitting loss using a preset loss function (such as the cross-entropy loss function). The fitting loss measures the deviation between the model's predictions and the actual labels. A smaller fitting loss indicates higher prediction accuracy; conversely, a larger fitting loss indicates lower prediction accuracy.

[0065] Model parameter adjustment: Determine whether the fitting loss is greater than a preset loss threshold. If the fitting loss is greater than the loss threshold, it indicates that the model's prediction accuracy does not meet the requirements. The parameters of the basic model (such as intercept, regression coefficient, etc.) need to be adjusted according to the magnitude and direction of the fitting loss. After adjustment, proceed to the step of substituting the multiple historical analysis vectors into the basic model to obtain multiple defect probability estimates, and recalculate the prediction and loss. If the fitting loss is not greater than the loss threshold, it indicates that the model's prediction accuracy meets the requirements, and no further parameter adjustment is needed.

[0066] Determine the defect analysis model: The basic model that has been adjusted and whose fitting loss meets the requirements is used as the final defect analysis model for cutting length analysis.

[0067] The basic model mentioned above adopts a logistic regression model, the expression of which is:

[0068] In the formula, This is a probability estimate, with a value ranging from [0,1]. is a natural constant, with a value of approximately 2.718. The intercept is... For the first The regression coefficients are used to characterize the degree of influence of the j-th feature on the defect probability. To analyze vectors, The total number of elements in the analysis vector is the dimension of the analysis vector.

[0069] For example, suppose 1000 historical analysis vectors are collected, each with dimension J=3, and the corresponding defect identifiers are known; substitute these historical analysis vectors into the basic model described above, with initial parameters... , , , 1000 defect probability estimates were obtained; the fitting loss was calculated using the cross-entropy loss function. If the loss value is 0.35, and the preset loss threshold is 0.1, the model parameters need to be adjusted, such as... Adjusted to 0.6. Adjust the value to 0.4 and recalculate the fitting loss until it drops below 0.1. The model at this point is the defect analysis model. Input the new second vector to be analyzed into the model. If the output defect probability estimate is 0.9 (the preset probability threshold is 0.85), it indicates that there is a defect in the area and it needs to be included in the clipping range.

[0070] In step 104, the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet are determined based on the first cutting length and the second cutting length.

[0071] In some embodiments, determining the tail end cutting length of the first billet and the head end cutting length of the second billet based on the first cutting length and the second cutting length includes: If the sum of the first cutting length and the second cutting length is greater than the sum of the first cutting lengths, then the tail end of the first steel billet is cut according to the first cutting length, and the head end of the second steel billet is cut according to the second cutting length. Otherwise, based on the first cutting length and the readjusted tail-end cutting length of the first billet and the head-end cutting length of the second billet, wherein the adjusted tail-end cutting length of the first billet is greater than the first cutting length, and the adjusted head-end cutting length of the second billet is greater than the second cutting length.

[0072] For example, the present invention uses the first cutting length determined in step 102 as a benchmark, and combines the first cutting length and the second cutting length obtained in step 103, and through a preset judgment logic, finally determines the tail cutting length of the first steel billet and the head cutting length of the second steel billet, so as to ensure that the cutting operation can completely remove the defects at the joint of the steel billet, and minimize steel waste, taking into account both product quality and production economy.

[0073] Based on the first cutting length and the second cutting length, the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet are determined, specifically including the following two cases: When the sum of the first cutting length and the second cutting length is greater than the sum of the first cutting lengths, it indicates that the total length of the actual defect area at the junction of the two steel billets exceeds the theoretical redundancy length. In this case, it is necessary to prioritize the complete removal of defects and cut according to the defect length obtained from the actual analysis. That is, cut the tail end of the first steel billet according to the first cutting length and cut the head end of the second steel billet according to the second cutting length to ensure that there are no defect residues at the ends of the steel billets after cutting.

[0074] When the sum of the first and second cutting lengths is not greater than the sum of the first cutting lengths, it indicates that the total length of the actual defect area at the junction of the two billets is less than the theoretical redundancy length. In this case, the tail-end cutting length of the first billet and the head-end cutting length of the second billet need to be readjusted based on the first cutting length. The core requirement for adjustment is that the tail-end cutting length of the first billet after adjustment must be greater than the first cutting length, and the head-end cutting length of the second billet after adjustment must be greater than the second cutting length. Furthermore, the sum of the two adjusted lengths must equal the sum of the first cutting lengths to avoid defect residue due to insufficient cutting. Simultaneously, it fully utilizes the theoretical redundancy length to ensure the flatness and quality stability of the billet ends after cutting. During adjustment, the cutting lengths can be allocated proportionally according to the defect distribution of the two billets, or a fixed adjustment ratio can be set based on production experience to ensure reasonable and efficient adjustment.

[0075] For example, suppose the first cutting length determined in step 102 is 1.6m, the first cutting length obtained in step 103 is 0.5m, and the second cutting length is 0.6m. The sum of the two is 1.1m, which is less than 1.6m. In this case, the cutting length needs to be adjusted. If the redundant length is allocated in a 1:1 ratio, the adjusted cutting length of the first billet tail end is 0.5 + (1.6m – 1.1m) / 2 = 0.75m, and the adjusted cutting length of the second billet head end is 0.6m + (1.6m – 1.1m) / 2 = 0.85m. The sum of the two after adjustment is 1.6m, and both are greater than their respective initial cutting lengths, which ensures defect removal and makes full use of the redundant length. If the first cutting length is 0.8m and the second cutting length is 0.9m, the sum of the two is 1.7m, which is greater than 1.6m. Then, the tail end of the first billet and the head end of the second billet are directly cut at 1.7m respectively to ensure complete removal of defects.

[0076] The method for adjusting the distance between the billet and the steel in this invention first obtains a first billet throwing time and a first billet biting time, wherein the first billet throwing time and the first billet biting time correspond to a first billet and a second billet, respectively, and the second billet enters the rolling mill production line after the first billet; then, based on the speed of the rolling mill production line, the first billet throwing time, and the first billet biting time, a first cutting length sum is determined, wherein the first cutting length sum is the sum of the tail cutting length of the first billet and the head cutting length of the second billet; then, based on the on-site process parameters of the rolling mill production line, a first current dataset, and a second current... The dataset is used to perform cutting length analysis to obtain a first cutting length and a second cutting length, wherein the first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. Finally, based on the first cutting length and the second cutting length, the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet are determined.

[0077] This invention provides a billet cutting method for a rolling mill production line, significantly improving cutting accuracy and ensuring product quality stability. By combining on-site process parameters and current datasets, this invention uses principal component analysis to remove redundancy and a defect analysis model to accurately identify defects, achieving precise location of defect areas at the billet ends and determining a reasonable cutting length. Through a dual-judgment logic, it balances defect removal and length benchmarks, effectively avoiding the problems of insufficient cutting (residual defects) or over-cutting (poor end flatness) caused by inaccurate defect identification and length calculation deviations in traditional cutting methods. This significantly improves the end quality of the cut billet, providing qualified billets for subsequent rolling processes and reducing product scrap rates due to cutting quality issues.

[0078] This invention provides a billet cutting method for a rolling mill production line, improving production efficiency and ensuring rolling continuity. The method clearly defines the billet connection sequence, ensuring precise connection between the second and first billets entering the production line, avoiding production interruptions. Scientific calculations determine the cutting length, providing a clear basis for the cutting operation and reducing waiting and adjustment time during the cutting process. Automated data analysis and defect identification processes replace traditional manual defect judgment, significantly shortening the time required for cutting length analysis and reducing subjective errors in manual judgment. Each step forms a closed-loop connection, effectively improving the automation level and operational efficiency of the cutting operation, ensuring continuous and stable operation of the rolling mill production line, and increasing overall production capacity.

[0079] This invention provides a billet cutting method for a rolling mill production line, ensuring production safety and reducing equipment wear. The maximum equipment reset time is carefully considered to ensure sufficient reset time for cutting-related equipment such as flying shears and loopers, preventing cutting operations from being performed before the equipment has fully reset, which could lead to equipment jamming, wear, or cutting deviations. This extends equipment lifespan and reduces maintenance costs. Simultaneously, automated cutting control reduces the frequency of close manual contact with equipment, lowering production safety hazards and improving the safe operation level of the production line.

[0080] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0081] The following are embodiments of the apparatus of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0082] Figure 3 This is a functional block diagram of the steel-following distance adjustment device provided in the embodiments of the present invention, with reference to... Figure 3 The steel-following distance adjustment device includes: a steel-following time acquisition module 301, a minimum cutting length acquisition module 302, a defect length determination module 303, and a cutting length determination module 304, wherein: The steel passing time acquisition module 301 is used to acquire the first steel throwing time and the first steel biting time, wherein the first steel throwing time and the first steel biting time correspond to the first steel billet and the second steel billet, respectively, and the second steel billet enters the rolling mill production line after the first steel billet; The minimum cutting length acquisition module 302 is used to determine the first cutting length sum based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, wherein the first cutting length sum is the sum of the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet; The defect length determination module 303 is used to perform cutting length analysis based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset to obtain the first cutting length and the second cutting length. The first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. The cutting length determination module 304 is used to determine the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet based on the first cutting length and the second cutting length.

[0083] Figure 4 This is a functional block diagram of the electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 of this embodiment includes a processor 400 and a memory 401, wherein the memory 401 stores a computer program 402 that can run on the processor 400. When the processor 400 executes the computer program 402, it implements the steps of the various steel-following distance adjustment methods and embodiments described above, for example... Figure 1 Steps 101 to 104 are shown.

[0084] For example, the computer program 402 may be divided into one or more modules / units, which are stored in the memory 401 and executed by the processor 400 to complete the present invention.

[0085] The electronic device 4 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The electronic device 4 may include, but is not limited to, a processor 400 and a memory 401. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0086] The processor 400 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0087] The memory 401 can be an internal storage unit of the electronic device 4, such as a hard disk or memory. The memory 401 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 401 can include both internal and external storage units of the electronic device 4. The memory 401 is used to store the computer program 402 and other programs and data required by the electronic device 4. The memory 401 can also be used to temporarily store data that has been output or will be output.

[0088] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0089] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0090] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0091] In the embodiments provided by this invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0092] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0093] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0094] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various methods and apparatus embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0095] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for adjusting the distance from the steel frame, characterized in that, include: The first steel throwing time and the first steel biting time are obtained, wherein the first steel throwing time and the first steel biting time correspond to the first steel billet and the second steel billet, respectively, and the second steel billet enters the rolling mill production line after the first steel billet; Based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, the first cutting length is determined, wherein the first cutting length is the sum of the tail cutting length of the first steel billet and the head cutting length of the second steel billet. Based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset, the cutting length is analyzed to obtain the first cutting length and the second cutting length. The first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. Based on the first cutting length and the second cutting length, the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet are determined.

2. The method for adjusting the distance between the steel beam and the steel frame according to claim 1, characterized in that, The determination of the first cutting length based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time includes: The difference between the first steel biting time and the first steel throwing time is taken as the steel passing interval time; The difference between the steel-passing interval time and the maximum equipment reset time is taken as the first interval time difference, wherein the maximum equipment reset time is the time determined based on the flying shear reset time and the maximum value among multiple looper reset times; Multiply the first time interval difference by the speed of the rolling mill production line to obtain the first cutting length.

3. The method for adjusting the distance between the steel beam and the steel frame according to claim 1, characterized in that, The step of performing cutting length analysis based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset to obtain the first cutting length and the second cutting length includes: For each current dataset in the first current dataset and the second current dataset, perform the following steps respectively: From the current dataset, extract the current parameter array of the rolling mill sequentially; The extracted current parameter array and the on-site process parameters are used to construct the first vector to be analyzed. The second vector to be analyzed is obtained by performing principal component analysis on the first vector to be analyzed using a dimensionality reduction matrix. The dimensionality reduction matrix is ​​obtained by performing principal component analysis on multiple historical vectors to be analyzed. The second vector to be analyzed is input into the defect analysis model to obtain the rolling defect probability; Add the obtained rolling defect probability to the defect probability queue; If the traversal of the current dataset is completed, the first cutting length or the second cutting length is determined according to the probability threshold and the defect probability queue. Otherwise, proceed to the step of sequentially extracting the current parameter array of the rolling mill from the current dataset.

4. The method for adjusting the distance between the steel beams according to claim 3, characterized in that, The dimensionality reduction matrix was obtained by performing principal component analysis on multiple historical vectors to be analyzed, including: Obtain the multiple historical vectors to be analyzed; The multiple historical vectors to be analyzed are standardized according to their dimensions to obtain multiple standard vectors; Perform cocorrelation analysis on the multiple standard vectors to obtain the cocorrelation matrix; Extract multiple eigenvalues ​​from the cocorrelation matrix; Multiple eigenvectors of the cocorrelation matrix are extracted based on the multiple eigenvalues, wherein each eigenvalue corresponds to one eigenvector; Multiple target values ​​are selected from the plurality of feature values ​​in descending order, wherein the ratio of the sum of the plurality of target values ​​to the sum of the plurality of feature values ​​is greater than the principal component threshold; Use the feature vector corresponding to the target value as the target vector; Multiple target vectors are used to construct the dimensionality reduction matrix.

5. The method for adjusting the distance between the steel beam and the steel frame according to claim 4, characterized in that, The standardization of the multiple historical vectors to be analyzed according to their dimensions yields multiple standard vectors, including: Perform the following steps for each dimension: Based on the dimensions, dimension data is extracted from the multiple historical vectors to be analyzed and constructed into a dimension array; Calculate the mean and standard deviation of multiple data points in the dimensional array; Based on the first formula, the mean of the multiple data points, and the standard deviation of the multiple data points, the dimensional array is standardized to obtain a standard array, wherein the first formula is: In the formula, For the first in the standard array One data point, For the 1st dimension in the array One data point, The mean of multiple data points. The standard deviation of multiple data points; Based on the correspondence with the dimension array, the data in the standard array is added to the standard vector.

6. The method for adjusting the distance between the steel beam and the steel frame according to claim 5, characterized in that, The defect analysis model is constructed based on multiple historical analysis vectors, including: Multiple historical analysis vectors are obtained, where each historical analysis vector corresponds to an identifier representing the existence of rolling defects. The historical analysis vectors are obtained by performing principal component analysis on the historical vectors to be analyzed using the dimensionality reduction matrix. Substitute the various historical analysis vectors into the basic model to obtain multiple defect probability estimates; The fitting loss is determined based on the multiple defect probability estimates and the identifiers of the multiple historical analysis vectors; If the fitting loss is greater than the loss threshold, the parameters of the basic model are adjusted according to the fitting loss, and the process jumps to the step of substituting the multiple historical analysis vectors into the basic model to obtain multiple probability values. Otherwise, the basic model shall be used as the defect analysis model; The basic model is as follows: In the formula, This is a probability estimate. It is a natural constant. The intercept is... For the first One regression coefficient, To analyze vectors, This is to analyze the total number of elements in the vector.

7. The method for adjusting the distance between the steel beam and the steel frame according to any one of claims 1-6, characterized in that, The step of determining the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet based on the first cutting length and the second cutting length includes: If the sum of the first cutting length and the second cutting length is greater than the sum of the first cutting lengths, then the tail end of the first steel billet is cut according to the first cutting length, and the head end of the second steel billet is cut according to the second cutting length. Otherwise, based on the first cutting length and the readjusted tail-end cutting length of the first billet and the head-end cutting length of the second billet, wherein the adjusted tail-end cutting length of the first billet is greater than the first cutting length, and the adjusted head-end cutting length of the second billet is greater than the second cutting length.

8. A device for adjusting the distance to steel, characterized in that, For implementing the steel follower distance adjustment method as described in any one of claims 1-7, the steel follower distance adjustment device comprises: The steel passing time acquisition module is used to acquire the first steel throwing time and the first steel biting time, wherein the first steel throwing time and the first steel biting time correspond to the first steel billet and the second steel billet, respectively, and the second steel billet enters the rolling mill production line after the first steel billet; The minimum cutting length acquisition module is used to determine the first cutting length sum based on the speed of the rolling mill production line, the first steel throwing time, and the first steel biting time, wherein the first cutting length sum is the sum of the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet; The defect length determination module is used to perform cutting length analysis based on the on-site process parameters of the rolling mill production line, the first current dataset, and the second current dataset to obtain the first cutting length and the second cutting length. The first cutting length and the second cutting length correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The first current dataset and the second current dataset correspond to the tail end of the first steel billet and the head end of the second steel billet, respectively. The current dataset includes an array of current parameters of the rolling mill collected at predetermined time intervals. as well as, The cutting length determination module is used to determine the tail end cutting length of the first steel billet and the head end cutting length of the second steel billet based on the first cutting length and the second cutting length.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7 above.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7 above.

Images