A method for determining impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes
By constructing a knowledge base for head impact compensation and working condition characteristics, the impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes were determined. This solved the problem of insufficient adaptability in parameter determination in existing technologies, achieved rapid and accurate wall thickness control, and improved the stability and consistency of production.
Patent Information
- Application Number
- CN202611053667.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-15
- Publication Date
- 2026-11-17
AI Technical Summary
In the control of the head wall thickness in the continuous rolling of seamless steel pipes, the existing technology relies on on-site operation experience and manual trial adjustment, which leads to insufficient adaptability and poor consistency in parameter determination, large fluctuations in head wall thickness, and difficulty in quickly and accurately determining impact compensation parameters.
A head impact compensation knowledge base is constructed. By collecting and structuring data from historical stable production batches, the head working condition characteristics of the target steel pipe are determined. An impact compensation benchmark is determined based on a set of similar historical batches. Parameters are corrected in conjunction with current working condition information, and equipment and process constraints are verified. Finally, the head impact compensation parameters are determined, and the parameters are updated and optimized through feedback.
It improves the accuracy and consistency of head wall thickness control, reduces reliance on manual experience and repeated trial and error, enhances the method's adaptability to different batches and operating conditions, ensures that parameters are within the feasible range, and improves production stability.
Smart Images

Figure CN123102758A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of seamless steel pipe continuous rolling control technology, and in particular to a method for determining impact compensation parameters for head wall thickness control in seamless steel pipe continuous rolling. Background Technology
[0002] During the continuous rolling process of seamless steel pipes, the pipe head is in an unsteady rolling stage when it enters the mill stand. It is susceptible to factors such as transient impacts, mill stand load fluctuations, rolling speed changes, temperature variations, and differences in incoming material conditions, making it difficult to control the pipe head wall thickness. Especially when specifications, steel grades, wall thickness targets, or production rhythms change, the pipe head wall thickness is more prone to deviations, affecting product quality stability and smooth production operations.
[0003] In existing seamless steel pipe continuous rolling mill head control, the determination of head compensation parameters largely relies on on-site operational experience, fixed experience tables, or manual trial adjustments to parameters such as main cylinder displacement and roll speed. This method has limited adaptability to different specifications, steel grades, and incoming material conditions, and the parameter determination process is highly dependent on operator experience, resulting in poor consistency across different shifts and operating conditions. When production conditions change, multiple trial adjustments are often required to approach the target state, easily leading to increased fluctuations in head wall thickness and higher debugging costs.
[0004] With the improvement of on-site testing and data acquisition capabilities, continuous rolling production can accumulate data such as specification information, steel grade information, process setting data, process operation data, head wall thickness detection results, and historical compensation parameter records. However, the current technology's utilization of this type of data is still relatively crude, lacking a method that can combine historical stable production batches and current continuous rolling conditions to collaboratively determine the main cylinder displacement compensation parameters and roll speed compensation parameters. Therefore, how to quickly and accurately determine the impact compensation parameters for head wall thickness control in seamless steel pipe continuous rolling has become an urgent technical problem to be solved in this field. Summary of the Invention
[0005] To address the technical problems in existing technologies where the determination of impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes relies primarily on on-site operational experience, fixed experience tables, or manual trial adjustments, resulting in insufficient adaptability, poor consistency, numerous trial adjustments, and large fluctuations in head wall thickness, this invention provides a method for determining impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes. The technical solution is as follows: A method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes, the method comprising: S1. Preset the head control range, collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data and historical compensation parameter records related to the continuous rolling control of the steel pipe head from historical stable production batches, and perform structured processing on the collected information and data to build a head impact compensation knowledge base. S2. Before the target steel pipe enters the continuous rolling head control stage, acquire the target steel pipe's specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control target, current continuous rolling working condition information, and head control interval information, and construct the target head working condition characteristics. S3. Based on the target head working condition characteristics, determine a set of similar historical batches in the head impact compensation knowledge base, and determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches; wherein, the head impact compensation benchmark includes the main cylinder displacement compensation benchmark value and the roll speed compensation benchmark value; S4. Based on the current continuous rolling load, rolling speed, temperature status, incoming material status and head wall thickness target of the target steel pipe, the head impact compensation benchmark is corrected to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe. S5. Perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters to determine the target steel pipe head impact compensation parameters. S6. Collect the head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, and update the head impact compensation knowledge base according to the head process feedback data and head wall thickness feedback data to correct the determination results of the head impact compensation parameters of subsequent steel pipes.
[0006] The historical compensation parameter records in S1 include historical master cylinder displacement compensation parameters and roll speed compensation parameter records.
[0007] The process of constructing the head impact compensation knowledge base in S1 is as follows: S11. Based on the timing information of the steel pipe head entering the continuous rolling mill stand and the preset head control interval, extract the head stage data related to head impact compensation from each historical stable production batch, and associate the head stage data, head wall thickness detection data and historical compensation parameter records with the steel pipe number, stand number and time sequence to form head impact compensation knowledge entries. The head stage data includes rolling force, rolling speed, main cylinder displacement response, roll speed response, stand load, temperature status, and tension status for each stand. S12. The head impact compensation knowledge entries are structured and stored to obtain a head impact compensation knowledge base: S121. Using the steel pipe number as the main association identifier of the head impact compensation knowledge entry, and the frame number and sampling time as the sub-association identifiers, the data corresponding to the same steel pipe are divided into the basic attribute field group, the head stage process field group, the compensation parameter field group and the wall thickness result field group, and written into the same head impact compensation knowledge entry. The basic attribute field group includes specification category, steel grade category, target wall thickness, and head control range; the head stage process field group includes rolling force, rolling speed, master cylinder displacement response, roll speed response, stand load, temperature status, and tension status for each stand; the compensation parameter field group includes historical master cylinder displacement compensation parameters and historical roll speed compensation parameters; and the wall thickness result field group includes average head wall thickness, maximum absolute head deviation, and head wall thickness deviation ratio. S122. Generate specification category label, steel grade category label, wall thickness target category label and continuous rolling condition category label according to the specification category, steel grade category, wall thickness target category label and continuous rolling condition category label in the head impact compensation knowledge entry, and form a composite retrieval index with the four category labels. Use the composite retrieval index to store and retrieve each head impact compensation knowledge entry to obtain the head impact compensation knowledge base. The continuous rolling condition category is a condition category label generated based on the head stage process field group, including high load type, medium load type, high speed type, low temperature type, or composite condition type.
[0008] Specifically, based on one or more of the following factors—stand load, rolling speed, temperature, and tension—the head rolling continuous rolling conditions are classified into high-load, medium-load, high-speed, low-temperature, or combined conditions.
[0009] Furthermore, feature values of stand load, rolling speed, temperature, and tension are extracted within the head control zone and compared with preset load thresholds, speed thresholds, temperature thresholds, and tension fluctuation thresholds. If only the stand load reaches the preset high load threshold, it is classified as a high load type; if the stand load is in the preset medium load range and does not meet the conditions for high speed, low temperature, or abnormal tension, it is classified as a medium load type; if only the rolling speed reaches the preset high speed threshold, it is classified as a high speed type; if only the temperature meets the preset low temperature threshold, it is classified as a low temperature type; if two or more conditions are met simultaneously, or if the tension is abnormal and accompanied by abnormal load, speed, or temperature, it is classified as a composite condition type. The relevant thresholds can be predetermined based on process specifications, equipment allowable ranges, or historical stable production batch statistics.
[0010] The specifications in S2 include outer diameter, wall thickness and length; the incoming material parameters include incoming material size and incoming material temperature; and the current continuous rolling condition information includes stand load, rolling speed, rotation speed status and temperature status. The specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control target, current continuous rolling condition information, and head control range information are combined and characterized to form the target head condition characteristics.
[0011] Specifically, S3 refers to: S31. Based on the target head working condition characteristics, extract candidate historical batches that match the target steel pipe in terms of specification category, steel type category, wall thickness target category and continuous rolling working condition category from the head impact compensation knowledge base. S32. A comprehensive evaluation is conducted on the differences in specifications, incoming materials, settings, working conditions, and target head wall thickness between the candidate historical batches and the target steel pipe. Based on the similarity evaluation value obtained from the comprehensive evaluation and the head wall thickness compliance results corresponding to each candidate historical batch, the candidate historical batches are screened and sorted to determine the set of similar historical batches. S33. Extract the historical compensation parameters corresponding to the set of similar historical batches (derived from the historical compensation parameter records associated with the head impact compensation knowledge entry), and determine the target steel pipe head impact compensation benchmark by combining the head wall thickness results of each similar historical batch.
[0012] The comprehensive evaluation in S32 is as follows: calculate the normalized difference between the candidate historical batch and the target steel pipe in each comparison field, and perform a weighted summation of each normalized difference to obtain the similarity evaluation value corresponding to each candidate historical batch. Filtering and sorting involve two layers of criteria: The first layer is the head wall thickness compliance screening. Based on the average deviation, maximum absolute deviation, and out-of-tolerance ratio of the head wall thickness of historical batches, it is determined whether the historical batch meets the preset wall thickness control requirements. Historical batches that do not meet the requirements are not used as the compensation benchmark.
[0013] The second layer is similarity ranking. Among the historical batches that meet the head wall thickness standard, they are ranked from smallest to largest according to the similarity evaluation value, and the top N batches are selected to form a set of similar historical batches.
[0014] Specifically, S4 refers to: S41. Based on the differences in continuous rolling load, incoming material condition, target head wall thickness, and rolling speed, as well as the correction coefficient of the main cylinder displacement compensation parameter, calculate the correction amount of the main cylinder displacement compensation. By adding the main cylinder displacement compensation reference value to the main cylinder displacement compensation correction amount, the corrected main cylinder displacement compensation parameter is obtained. S42. Based on the differences in rolling speed, temperature conditions, continuous rolling load, and target head wall thickness, as well as the correction coefficient of the roll speed compensation parameter, calculate the roll speed compensation correction amount. By adding the roll speed compensation reference value to the roll speed compensation correction amount, the corrected roll speed compensation parameter is obtained. S43. Combining the coordinated matching relationship between the modified main cylinder displacement compensation parameters and the modified roll speed compensation parameters, and the target head wall thickness of the target steel pipe, the head impact compensation parameters of the target steel pipe are formed; wherein, the head impact compensation parameters of the target steel pipe include the main cylinder displacement compensation parameters and the roll speed compensation parameters.
[0015] Specifically, S43 refers to: The normalized change of the main cylinder displacement compensation parameter relative to its reference value and the normalized change of the roll speed compensation parameter relative to its reference value are calculated respectively, and the coordination deviation between the normalized changes of the two is calculated according to the preset coordination coefficient. When the coordination deviation does not exceed the preset coordination threshold, the two compensation parameters are considered to be matched. When the coordination deviation exceeds the preset threshold, at least one of the parameters is coordinated and corrected.
[0016] When the amount of the coordination deviation is not greater than the preset coordination threshold, the corrected main cylinder displacement compensation parameter and the corrected roll speed compensation parameter are combined to form a head impact compensation parameter pair. When the amount of the coordination deviation is greater than the preset coordination threshold, the adjustable margins of the main cylinder displacement compensation parameter and the roll speed compensation parameter from the corresponding equipment upper and lower limits are calculated respectively, and the compensation parameter with the larger adjustable margin is selected as the parameter to be adjusted. According to the preset correction step size, the parameter to be adjusted is increased or decreased successively in the direction that reduces the amount of cooperative deviation, and the amount of cooperative deviation is recalculated after each adjustment. When the parameter to be adjusted reaches the allowable boundary of the corresponding device and the coordination deviation is still greater than the preset coordination threshold, switch to another compensation parameter and continue to adjust it step by step until the coordination deviation is no greater than the preset coordination threshold. When the two modified parameters meet the requirements of two-parameter coordinated matching, they are combined to form the head impact compensation parameters: Head impact compensation parameters = {main cylinder displacement compensation parameters, roll speed compensation parameters}.
[0017] When the two do not meet the matching requirements, at least one of the parameters is re-coordinated and corrected until the matching threshold is met, and then the two corrected parameters are combined as the head impact compensation parameters.
[0018] Specifically, S5 is: S51. Extract the equipment constraint information and process constraint information corresponding to the continuous rolling control of the target steel pipe head; The equipment constraint information includes the allowable range of main cylinder displacement, the allowable range of main cylinder adjustment rate, the allowable range of roll speed, the allowable range of speed variation, and the frame load limit. The process constraint information includes the head wall thickness control range, the head wall thickness allowable deviation, and the compensation parameter collaborative matching threshold. S52. Compare the main cylinder displacement compensation parameters and roll speed compensation parameters obtained in S4 with the equipment constraint information and process constraint information in S51 respectively. Correct the compensation parameters that exceed the constraint range within the constraint range. Then, judge the feasibility and consistency of the head wall thickness target for the corrected candidate head impact compensation parameters to determine the final head impact compensation parameters of the target steel pipe.
[0019] The constraint correction in S52 specifically refers to: Parameters exceeding the allowable range are truncated to the corresponding boundary values, and the parameters are graded and reverted according to the master cylinder adjustment rate, speed change, and frame load limit.
[0020] The feasibility judgment in S52 refers to whether the candidate head impact compensation parameters simultaneously meet the equipment execution boundaries, including: the allowable range of main cylinder displacement, the allowable range of main cylinder adjustment rate, the allowable range of roll speed, the allowable range of speed change, the upper limit of frame load, and the dual-parameter collaborative deviation threshold; when all are met, the candidate parameters are judged to be feasible.
[0021] The consistency judgment of the head wall thickness target in S52 refers to whether the predicted head wall thickness value estimated based on the candidate master cylinder displacement compensation parameters and roll speed compensation parameters falls within the control range formed by the head wall thickness target value and the allowable deviation.
[0022] In step S52, when a candidate head impact compensation parameter satisfies the equipment constraints and the corresponding head wall thickness prediction value is within the head wall thickness control range, it is determined as the final head impact compensation parameter; otherwise, multiple correction parameter groups are generated in the neighborhood of the candidate parameter, and the parameter group that satisfies the constraints and has the smallest head wall thickness prediction deviation is selected as the final head impact compensation parameter.
[0023] The update process in S6 is as follows: S61. Based on the difference between the measured wall thickness value of each sampling point in the head control interval and the target value of the head wall thickness, calculate the average deviation of the head wall thickness, the maximum absolute deviation of the head wall thickness, and the proportion of head wall thickness exceeding the tolerance, which are used as head wall thickness control deviation indicators. S62. Write the target head working condition characteristics, final head impact compensation parameters, head process feedback data, head wall thickness feedback data and head wall thickness control deviation index into the same head impact compensation knowledge entry to form an updated head impact compensation knowledge entry and update the head impact compensation knowledge base. S63. Based on the head wall thickness control deviation index and the head process feedback data, when any one of the head wall thickness control deviation indices exceeds the corresponding preset threshold, a correction is triggered; the selection basis for the compensation benchmark, parameter correction rules, or constraint correction basis used when determining the subsequent steel pipe head impact compensation parameters are corrected; if the threshold is not exceeded, only the production data of this production is written to the knowledge base.
[0024] The apparatus for implementing the method includes: The knowledge base construction unit is used to collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data, and historical compensation parameter records related to the continuous rolling control of the steel pipe head from historical stable production batches, and to perform structured processing on the data to construct a head impact compensation knowledge base; The working condition characterization unit is used to acquire the specification information, head wall thickness control target and current continuous rolling working condition information of the target steel pipe before the target steel pipe enters the continuous rolling head control stage, and to construct the target head working condition characteristics. The benchmark determination unit is used to determine a set of similar historical batches in the head impact compensation knowledge base according to the working condition characteristics of the target head, and to determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches. The parameter correction unit is used to correct the head impact compensation benchmark by combining the current continuous rolling load, rolling speed, temperature state, incoming material state and head wall thickness target of the target steel pipe, so as to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe. The parameter determination unit is used to perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters, and to determine the target steel pipe head impact compensation parameters. The feedback update unit is used to collect head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, update the head impact compensation knowledge base according to the process feedback data and the head wall thickness feedback data, and correct the determination results of head impact compensation parameters for subsequent steel pipes.
[0025] The present invention also provides an electronic device, the electronic device comprising: a processor; a memory storing computer-readable instructions, wherein when the computer-readable instructions are executed by the processor, any one of the methods described above for determining the impact compensation parameters for controlling the head wall thickness of the continuous rolling mill of seamless steel pipes is implemented.
[0026] The present invention provides a computer-readable storage medium storing program code, which can be called by a processor to execute any of the methods described above for determining the impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes.
[0027] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this invention, by constructing a head impact compensation knowledge base, the specification information, process setting data, process operation data, head wall thickness detection results, and historical compensation parameter records related to head wall thickness control in historical stable production batches are structured and organized, so that historical effective production experience can be accumulated and reused, providing a data basis for determining the head impact compensation parameters of subsequent target steel pipes. By constructing target head working condition characteristics and filtering similar historical batch sets from the head impact compensation knowledge base, the head impact compensation benchmark can be determined based on historical effective experience, enhancing the pertinence and reusability of the compensation parameter determination process. By combining the current continuous rolling load, rolling speed, temperature, incoming material condition and head wall thickness target to correct the impact compensation benchmark, and by coordinating the determination of the main cylinder displacement compensation parameters and the roll speed compensation parameters, the determined impact compensation parameters can be more in line with the current working conditions, thereby improving the accuracy and consistency of head wall thickness control. By introducing equipment constraint verification and process constraint verification, it is possible to avoid the determined impact compensation parameters from exceeding the feasible range or not matching the head wall thickness control requirements, thereby improving the feasibility of impact compensation parameters and the stability of head wall thickness control. By updating the head impact compensation knowledge base based on head process feedback data and head wall thickness feedback data, and correcting the subsequent determination results of impact compensation parameters for steel pipes, the method can improve its adaptability to different batches and working conditions, and reduce reliance on manual experience and repeated trial and error. Attached Figure Description
[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.
[0029] Figure 1 This is a flowchart of a method for determining impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes, provided by an embodiment of the present invention. Figure 2 This is a flowchart illustrating the overall process of determining impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes, as provided in this embodiment of the invention. Figure 3 This is a flowchart illustrating the construction of a head impact compensation knowledge base, the construction of target head working condition features, the screening of similar historical batches, and the determination of head impact compensation benchmarks provided in this embodiment of the invention. Figure 4This is a flowchart illustrating the correction, constraint verification, and feedback update of main cylinder displacement compensation parameters and roll speed compensation parameters provided in an embodiment of the present invention. Figure 5 This is a schematic diagram of the structure of a device and electronic equipment for determining the impact compensation parameters for the head wall thickness control of a seamless steel pipe in continuous rolling mill, provided in an embodiment of the present invention. Detailed Implementation
[0030] The technical solution of the present invention will now be described with reference to the accompanying drawings.
[0031] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.
[0032] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.
[0033] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.
[0034] This invention provides a method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes. This method can be implemented using a device for determining these parameters. The device can be an edge computing node of the continuous rolling production line, a factory server, or a cloud computing platform, capable of interacting with the production process data acquisition system, the head wall thickness detection system, and a historical knowledge base. Figure 2 The flowchart shown is an overall process for determining the impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes. In this embodiment of the invention, historical stable production batch data, target steel pipe operating condition information, and head wall thickness control target are used as inputs. The process sequentially executes knowledge base construction, target operating condition characterization, similar historical batch screening, compensation benchmark determination, parameter correction, constraint verification, and feedback update to finally obtain the head impact compensation parameters for the target steel pipe. Figure 1 The method may include the following steps: S1. Preset the head control range, collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data and historical compensation parameter records related to the continuous rolling control of the steel pipe head from historical stable production batches, and perform structured processing on the collected information and data to build a head impact compensation knowledge base. S2. Before the target steel pipe enters the continuous rolling head control stage, acquire the target steel pipe's specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control target, current continuous rolling working condition information, and head control interval information, and construct the target head working condition characteristics. S3. Based on the target head working condition characteristics, determine a set of similar historical batches in the head impact compensation knowledge base, and determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches; wherein, the head impact compensation benchmark includes the main cylinder displacement compensation benchmark value and the roll speed compensation benchmark value; S4. Based on the current continuous rolling load, rolling speed, temperature status, incoming material status and head wall thickness target of the target steel pipe, the head impact compensation benchmark is corrected to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe. S5. Perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters to determine the target steel pipe head impact compensation parameters. S6. Collect the head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, and update the head impact compensation knowledge base according to the head process feedback data and head wall thickness feedback data to correct the determination results of the head impact compensation parameters of subsequent steel pipes.
[0035] The specific implementation process will be explained below.
[0036] like Figure 3 Steps S1 to S3 are mainly used to complete the construction of the head impact compensation knowledge base, the construction of target head working condition features, the screening of similar historical batches, and the determination of the head impact compensation benchmark, providing a basis for subsequent compensation parameter correction.
[0037] S1. Collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data, and historical compensation parameter records related to the continuous rolling control of the steel pipe head from historical stable production batches, and perform structured processing on the data to construct a head impact compensation knowledge base.
[0038] In one feasible implementation, the purpose of this step is to extract basic data, process data, and result data that can characterize the wall thickness control status and compensation effect of the steel pipe head from historical stable production batches, and organize them into searchable, associative, and reusable knowledge entries to form a head impact compensation knowledge base required for subsequent screening of similar historical batches and determination of head impact compensation benchmarks.
[0039] Optionally, step S1 above may include the following steps S11 to S12: S11. Collect the specification information, steel grade information, incoming material size information, continuous rolling setting parameters, stand process operation parameters, head wall thickness detection results, and historical main cylinder displacement compensation parameters and roll speed compensation parameters corresponding to historical stable production batches.
[0040] In one feasible implementation, historical batches that meet stable production conditions are first selected from the historical production database. These stable production conditions may include: no abnormal shutdowns during production, critical rack load fluctuations not exceeding a preset threshold, head wall thickness detection results meeting a preset control range, and complete production records.
[0041] For ease of representation, the set of historically stable production batches is defined as: (1) In the formula, B represents the set of historically stable production batches; b u Indicates the first u A number of historically stable production batches; N B This represents the total number of historically stable production batches; u This represents an index of historically stable production batches.
[0042] Furthermore, for any historically stable production batch b u Its collection of fields can be represented as: (2) In the formula, D u Indicates the first u A collection of data fields for a historically stable production batch; g u Indicates specification information; s u Indicates steel type information; m u Indicates the dimensions of the incoming material; q u This indicates the continuous rolling mill setting parameters; x u Indicates the rack process operating parameters; w u This indicates the head wall thickness test results; p cyl,u This indicates the historical master cylinder displacement compensation parameter records; p spd,u This represents the historical roll speed compensation parameter record.
[0043] In one feasible implementation, the specification information includes at least the finished product outer diameter, target wall thickness, and steel pipe length; the incoming material information includes at least the incoming material outer diameter and incoming material wall thickness; the continuous rolling setting parameters include at least the stand setting speed, setting reduction, and setting tension; the stand process operation parameters include at least the rolling force, rolling speed, main cylinder displacement response, and roll speed response; and the head wall thickness detection results include at least the head average wall thickness, head maximum deviation, and head minimum deviation.
[0044] Table 1 provides examples of some collected data from historically stable production batches.
[0045] Table 1. Examples of partial data collected from historical stable production batches
[0046] S12. The head impact compensation knowledge entries are structured and stored to obtain a head impact compensation knowledge base: S121. Based on the timing information of the steel pipe head entering the continuous rolling mill stand and the preset head control range, extract the head stage data related to head impact compensation from each historical stable production batch, and associate the head stage data, head wall thickness detection results and historical compensation parameter records with the steel pipe number, stand number and time sequence to form head impact compensation knowledge entries.
[0047] Generally, the steel pipe number is used as the main association identifier for the head impact compensation knowledge entry, and the frame number and sampling time are used as the sub-association identifiers. The data corresponding to the same steel pipe are divided into basic attribute field group, head stage process field group, compensation parameter field group and wall thickness result field group, and written into the same head impact compensation knowledge entry. In one feasible implementation, based on historical stable production batches, the head control stage is further determined according to the timing information of the steel pipe head entering the continuous rolling mill stand. The head control stage can be determined jointly by the moment the head enters the first continuous rolling mill stand and a preset head control length range.
[0048] Definition of the first u The head control time interval and head control length interval for each historically stable production batch are as follows: (3) (4) In the formula, T u Indicates the first u The head control time interval of a historically stable production batch, L u Indicates the first u The head control length range of a historically stable production batch. and These represent the start and end times of head control, respectively. and These represent the start and end points of the head control length, respectively.
[0049] In one feasible implementation, the preset head control length range can be set to 0.0m to 2.5m after the beginning of the steel pipe head based on field experience; different head control length ranges can also be used for steel pipes of different specifications.
[0050] Table 2 provides examples of head control intervals for some historically stable production batches.
[0051] Table 2 Examples of Head Control Ranges for Historically Stable Production Batches
[0052] After extracting the head stage data, it is necessary to correlate the head stage data, head wall thickness detection results, and historical compensation parameter records accordingly. For this purpose, the following is defined: u The knowledge item for head impact compensation corresponding to each historically stable production batch is: (5) In the formula, Indicates the first u Knowledge items related to head impact compensation for each historically stable production batch. Represents a subset of basic attributes; This represents a subset of the head stage process data; This represents a subset of the head wall thickness results.
[0053] In one feasible implementation, the basic attribute subset includes at least the pipe number, specification category, steel type, and target wall thickness; the head stage process data subset includes at least the rolling force of each stand, rolling speed, main cylinder displacement response, and roll speed response; and the head wall thickness result subset includes at least the head average wall thickness, head maximum deviation, and head minimum deviation.
[0054] S122. The head impact compensation knowledge entries are stored in a structured manner to construct a head impact compensation knowledge base.
[0055] Generally, specification category tags, steel grade tags, wall thickness target category tags, and continuous rolling condition category tags are generated according to the specification category, steel grade category, wall thickness target category tag, and continuous rolling condition category in the head impact compensation knowledge entries, and the four category tags are combined into a composite retrieval index. The composite retrieval index is used to store and retrieve each head impact compensation knowledge entry to obtain the head impact compensation knowledge base. In one feasible implementation, head impact compensation knowledge entries formed from each historical stable production batch are stored according to a unified field structure, and an index field is established for subsequent retrieval and matching, thereby forming a head impact compensation knowledge base.
[0056] Define the head impact compensation knowledge base as follows: (6) Furthermore, to improve the efficiency of knowledge item retrieval and filtering, the first... u The index field group for each knowledge item is: (7) In the formula, I u Indicates the first u A group of index fields for each knowledge entry; Indicates the specification category identifier; Indicates the steel grade category; Indicates the target wall thickness; c u This indicates the category of continuous rolling operation.
[0057] In one feasible implementation, the continuous rolling condition category is determined as follows: within the head control zone, the stand load characteristic value, rolling speed characteristic value, temperature state characteristic value, and tension state characteristic value are extracted respectively; wherein, the stand load characteristic value can be represented by the average, maximum, or weighted average of the load rates of each stand within the head control zone; the rolling speed characteristic value can be represented by the average rolling speed within the head control zone or the deviation relative to the set speed; the temperature state characteristic value can be represented by the average of the incoming material temperature or the rolling temperature within the head control zone; and the tension state characteristic value can be represented by the tension fluctuation amplitude, tension standard deviation, or deviation relative to the set tension within the head control zone.
[0058] When the stand load characteristic value is greater than or equal to a preset high load threshold, and the rolling speed characteristic value, temperature state characteristic value, and tension state characteristic value do not simultaneously meet other abnormal operating condition judgment conditions, the head continuous rolling condition is classified as a high load type; when the stand load characteristic value is in a preset medium load range, and the rolling speed characteristic value does not reach a preset high speed threshold, the temperature state characteristic value is not lower than a preset low temperature threshold, and the tension state characteristic value does not exceed a preset tension fluctuation threshold, the head continuous rolling condition is classified as a medium load type; when the rolling speed characteristic value is greater than or equal to a preset high speed threshold, and does not simultaneously meet the high load or low temperature operating condition judgment conditions, the head continuous rolling condition is classified as a high speed type; when the temperature state characteristic value is less than or equal to a preset low temperature threshold, and does not simultaneously meet the high load or high speed operating condition judgment conditions, the head continuous rolling condition is classified as a low temperature type; when two or more of the judgment conditions of high load, high speed, low temperature, and tension fluctuation abnormality are simultaneously met, the head continuous rolling condition is classified as a composite operating condition type.
[0059] The preset high load threshold, preset medium load range, preset high speed threshold, preset low temperature threshold, and preset tension fluctuation threshold can be predetermined based on the continuous rolling process specifications, equipment allowable range, statistical distribution of historical stable production batches, or field experience parameters.
[0060] Table 3 provides a partial example of the structured storage of knowledge entries related to head impact compensation.
[0061] Table 3. Example of structured storage for knowledge items related to head impact compensation.
[0062] Therefore, by collecting basic and process data corresponding to historical stable production batches, extracting head stage data, forming head impact compensation knowledge entries and storing them in a structured manner, the head impact compensation knowledge base was constructed, providing a data foundation for the subsequent construction of target steel pipe head working condition characteristics, screening of similar historical batches, and determination of head impact compensation parameters.
[0063] S2. Before the target steel pipe enters the continuous rolling head control stage, obtain the specification information, head wall thickness control target and current continuous rolling condition information of the target steel pipe, and construct the target head condition characteristics.
[0064] In one feasible implementation, the purpose of this step is to obtain the basic parameters, head wall thickness control requirements and current continuous rolling conditions of the target steel pipe before entering the continuous rolling head control stage, and to organize the information into target head condition features that can be used for retrieval and comparison with historical stable production batches, so as to support the determination of subsequent similar historical batch sets and the formation of head impact compensation benchmarks.
[0065] Optionally, step S2 above may include the following steps S21-S22: S21. Obtain the target steel pipe's corresponding specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control requirements, current working condition information, and head control range information.
[0066] In one feasible implementation, the basic parameters of the target steel pipe are first obtained. These basic parameters include at least specification parameters, steel grade parameters, incoming material parameters, and continuous rolling setting parameters; wherein, the specification parameters include at least the finished product outer diameter, target wall thickness, and target length; the steel grade parameters include at least the steel grade category or material identifier; the incoming material parameters include at least the incoming material outer diameter, incoming material wall thickness, and incoming material temperature; and the continuous rolling setting parameters include at least one or more of the following: set speed, set reduction, and set tension.
[0067] Furthermore, the head wall thickness control requirements for the target steel pipe are obtained. These head wall thickness control requirements include at least a target head wall thickness and an allowable deviation in head wall thickness; wherein, the target head wall thickness can be represented by a single target value, and the allowable deviation in head wall thickness can be represented by an allowable fluctuation range corresponding to the target value; in another embodiment, it can also be directly represented by an upper limit and a lower limit interval for head wall thickness.
[0068] Furthermore, the current operating condition information of the target steel pipe during the head control stage is obtained. The current operating condition information includes at least the stand load, rolling speed, rotational speed, and temperature, and the corresponding acquisition time or acquisition interval is recorded; in one feasible implementation, the current operating condition information is characterized by the mean, extreme value, or steady-state representative value within the head control interval.
[0069] For ease of representation, the original head condition information set of the target steel pipe is defined as follows: (8) In the formula, This represents the set of original head working condition information for the target steel pipe; g Indicates specifications and parameters; s Indicates steel grade parameters; m Indicates incoming material parameters; q This indicates the continuous rolling mill setting parameters; w This indicates the head wall thickness control requirements; x Indicates current operating status information; This indicates the header control interval information.
[0070] Table 4 provides some example data on the original working conditions of the target steel pipe head.
[0071] Table 4 Example of original head working condition information of the target steel pipe
[0072] S22. The specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control requirements, current working condition information, and head control range information are combined and characterized to form the target head working condition characteristics.
[0073] In one feasible implementation, after obtaining the original head condition information set of the target steel pipe, instead of simply stacking the parameters, the fields used for subsequent comparison and retrieval are organized into comparable objects. In another feasible implementation, the comparable objects include two parts: one part is the original head condition information set obtained in step S21. The other part is a set of working condition tags for quick filtering. .
[0074] For ease of representation, the target head condition feature is defined as follows: (9) In the formula, Indicates the target head's working condition characteristics; This represents a set of operating condition labels.
[0075] The set of operating condition labels can be represented as: (10) In the formula, g tag Indicates the specification category of the target steel pipe; s tag Indicates the steel type of the target steel pipe; w tag Indicates the target wall thickness of the target steel pipe; c This indicates the continuous rolling condition category of the target steel pipe.
[0076] In one feasible implementation, the original head condition information set At a minimum, it should include the target steel pipe's specifications, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control requirements, and numerical fields from the current operating condition information; a set of operating condition tags. It includes at least specification category labels, steel grade category labels, target wall thickness labels, and continuous rolling condition category labels. This allows the target header condition features to retain both basic parameters and detailed information about the current condition, while also forming a unified object that can be quickly filtered and compared with historical knowledge entries.
[0077] Table 5 provides some examples of the target head's working condition characteristics.
[0078] Table 5 Examples of target head working condition characteristics
[0079] Therefore, by acquiring the basic parameters of the target steel pipe, the head wall thickness control requirements, the current working condition information, and the head control range information, and organizing them into a set of original head working condition information, the information can be obtained. and working condition tag set This formed the target head working condition characteristics. This provides an input basis for determining the subsequent set of similar historical batches.
[0080] S3. Based on the target head working condition characteristics, determine the set of similar historical batches in the head impact compensation knowledge base, and determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches.
[0081] In one feasible implementation, the purpose of this step is to select historically stable production batches that are similar to the target steel pipe head control state from the head impact compensation knowledge base constructed in step S1, based on the target head working condition characteristics formed in step S2, and to determine the target steel pipe head impact compensation benchmark on this basis. The head impact compensation benchmark includes the benchmark value of the main cylinder displacement compensation parameter and the benchmark value of the roll speed compensation.
[0082] Optionally, step S3 above may include the following steps S31-S33: S31. Based on the target head working condition characteristics, extract candidate historical batches from the head impact compensation knowledge base that match the target steel pipe in terms of specification category, steel type category, wall thickness target category, and continuous rolling working condition category.
[0083] In one feasible implementation, to avoid blindly comparing all historical entries in the head impact compensation knowledge base, the first step is to base the comparison on the set of working condition labels in the target head's working condition characteristics. A preliminary screening of historical knowledge entries is conducted. The preliminary screening dimensions include at least specification category, steel grade category, target wall thickness category, and continuous rolling condition category.
[0084] Definition of the first u The initial screening index for each historical knowledge item is: (11) In the formula, χ u Indicates the first u The initial screening index for each historical knowledge item; 1 ( The parentheses indicate an indicator function that takes the value 1 if the condition inside the parentheses is true, and 0 otherwise.
[0085] Furthermore, the candidate historical batch set is defined as: (12) In the formula, This represents the set of candidate historical batches.
[0086] In one feasible implementation, the purpose of the initial screening is to remove obviously irrelevant historical batches, such as those with inconsistent specifications, steel grades, wall thickness targets, or continuous rolling conditions, so that they will not be included in the subsequent similarity evaluation process, thereby reducing the amount of calculation required for subsequent comparisons and improving the targeting of the screening.
[0087] Table 6 provides some example results of the initial screening of candidate historical batches.
[0088] Table 6. Examples of preliminary screening results for candidate historical batches
[0089] S32. A comprehensive evaluation is conducted on the differences in specifications, incoming materials, settings, working conditions, and head wall thickness between the candidate historical batches and the target steel pipe to obtain the similarity evaluation value corresponding to each candidate historical batch. Based on the similarity evaluation value and the head wall thickness compliance results corresponding to each candidate historical batch, the candidate historical batches are screened and sorted to determine the set of similar historical batches.
[0090] In one feasible implementation, in the candidate historical batch set Once formed, a comprehensive evaluation is performed on the differences between each candidate historical batch and the target steel pipe. These differences include at least specification differences, incoming material differences, setting differences, operating condition differences, and head wall thickness target differences. To eliminate the influence of different dimensions, various differences can be normalized before a comprehensive score is calculated.
[0091] Definition of the first u The normalized differences of the candidate historical batches across various fields are as follows: (13) In the formula, Indicates the first u The candidate historical batch in the first k Normalized differences on class fields Indicates the first u The candidate historical batch in the first k The values on the comparison fields are compared. Indicates the target steel pipe in the first... k The values on the comparison fields are compared. Indicates the first k The normalization scale corresponding to the class field; where, k =1,2,3,4,5 correspond to specification differences, incoming material differences, setting differences, working condition differences, and head wall thickness target differences, respectively.
[0092] Furthermore, define the first u The similarity evaluation value for each candidate historical batch is: (14) In the formula, r u Indicates the first u The similarity evaluation value of each candidate historical batch. α (k) Indicates the first k The weight coefficients corresponding to class differences, and satisfying: (15) In one feasible implementation, the smaller the similarity evaluation value, the closer the candidate historical batch is to the head working condition of the target steel pipe.
[0093] Furthermore, to avoid mistakenly selecting batches that, while similar, have historically exhibited poor head wall thickness control as reference samples, it is necessary to combine the head wall thickness compliance results for judgment. Definition of the first... u The wall thickness compliance indicator for each candidate historical batch is: (16) In the formula, π u Indicates the first u The wall thickness compliance indicator for each candidate historical batch. Indicates the first u Average deviation of head wall thickness in candidate historical batches Indicates the first u The maximum absolute deviation of the head wall thickness of each candidate historical batch. Indicates the first u The percentage of out-of-tolerance wall thickness at the head of each candidate historical batch. , and These represent the corresponding upper limit thresholds.
[0094] Define the set of similar historical batches as: (17) In the formula, Represents a set of similar historical batches, Top N ( This indicates that among the candidate historical batches that meet the wall thickness standard, they are sorted by similarity evaluation value from smallest to largest, and the top batches are retained. N Each batch is used as a set of similar historical batches. N This indicates the number of similar historical batches retained.
[0095] Table 7 provides some examples of similarity evaluation and wall thickness compliance results.
[0096] Table 7 Examples of similarity evaluation and wall thickness compliance results for candidate historical batches
[0097] S33. Extract the historical compensation parameters corresponding to the set of similar historical batches from the historical compensation parameter records associated with the head impact compensation knowledge entries, and determine the target steel pipe head impact compensation benchmark by combining the head wall thickness results of each similar historical batch.
[0098] In one feasible implementation, after obtaining a set of similar historical batches, it is necessary to determine the head impact compensation benchmark for the target steel pipe based on the historical master cylinder displacement compensation parameters and historical roll speed compensation parameters corresponding to the similar historical batches. In another feasible implementation, if there is only one optimal batch in the set of similar historical batches, the historical compensation parameters of that optimal batch can be directly used as the head impact compensation benchmark. In this embodiment, the head impact compensation benchmark is formed by weighting the historical compensation parameters of multiple similar historical batches, and is further corrected by combining the head wall thickness deviation of each similar historical batch. The head wall thickness deviation is used to characterize the quality of the head wall thickness control effect of the similar historical batches, and can be obtained from the average head wall thickness deviation, the maximum absolute deviation, or a combination thereof.
[0099] Definition of the first u The overall weight of similar historical batches is: (18) In the formula, ρ u Indicates the first u The overall weight of each similar historical batch is calculated, and the sum of the overall weights of all similar historical batches is 1. ε This represents a positive constant to prevent the denominator from being zero.
[0100] Furthermore, the impact compensation benchmark for the target steel pipe head is defined as follows: (19) In the formula, This indicates the impact compensation benchmark for the head of the target steel pipe. This indicates the reference value for master cylinder displacement compensation. This indicates the reference value for roll speed compensation.
[0101] in, (20) Table 8 provides some examples of similar historical batch sets and the determination of head impact compensation benchmarks.
[0102] Table 8. Examples of similar historical batch sets and head impact compensation benchmarks
[0103] In the above example, after weighted calculation using equation (20), the target steel pipe's main cylinder displacement compensation benchmark value is 1.73 mm, and the roll speed compensation benchmark value is 5.77 r / min. Thus, the process of determining the head impact compensation benchmark from a set of similar historical batches is completed.
[0104] Therefore, by conducting preliminary screening, similarity evaluation, and standard screening of historical knowledge entries in the head impact compensation knowledge base based on the target head working condition characteristics, a set of similar historical batches is formed, and the target steel pipe head impact compensation benchmark is further determined, providing an input basis for subsequent head impact compensation parameter correction.
[0105] like Figure 4 As shown, steps S4 to S6 are mainly used to correct the main cylinder displacement compensation parameters and roll speed compensation parameters, verify equipment and process constraints, and update the knowledge base based on the head impact compensation benchmark.
[0106] S4. Based on the current continuous rolling load, rolling speed, temperature, incoming material condition, and head wall thickness target of the target steel pipe, the head impact compensation benchmark is corrected to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe.
[0107] In one feasible implementation, the purpose of this step is to modify the main cylinder displacement compensation benchmark value and the roll speed compensation benchmark value in the head impact compensation benchmark based on the target steel pipe head impact compensation benchmark obtained in step S3, taking into account the difference between the current working condition of the target steel pipe and the reference state of similar historical batches, and further form head impact compensation parameters for target steel pipe head wall thickness control through dual-parameter collaborative matching relationship.
[0108] Optionally, step S4 above may include the following steps S41-S43: First, extract the difference information of the target steel pipe relative to the similar historical batch set in terms of continuous rolling load, rolling speed, temperature state, incoming material state and head wall thickness target, and form the basis for head impact compensation correction.
[0109] In one feasible implementation, the relevant status information of the target steel pipe is first compared with the reference status corresponding to the set of similar historical batches determined in step S3, and the difference information used for subsequent parameter correction is extracted. The difference information includes at least: continuous rolling load difference, rolling speed difference, temperature condition difference, incoming material condition difference, and head wall thickness target difference.
[0110] In one feasible implementation, the reference values of similar historical batch sets across various fields are obtained using a weighted average method, namely: (twenty one) In the formula, The set of similar historical batches is represented in the first... k Reference values on class fields Indicates the first u The first similar historical batch in k The values that can be obtained on a class field; where, k =1,2,3,4,5 correspond to continuous rolling load, rolling speed, temperature condition, incoming material condition, and head wall thickness target, respectively.
[0111] Furthermore, to eliminate the influence of different dimensions of the differences, the differences are normalized, resulting in: (twenty two) In the formula, Δ (1) Δ (2) Δ (3) Δ (4) and Δ (5) These represent differences in continuous rolling load, rolling speed, temperature condition, incoming material condition, and target head wall thickness, respectively. y (k) Indicates the target steel pipe in the first... k Values that can be retrieved on class fields C (k) Indicates the first k The normalization scale corresponding to the class field.
[0112] Table 9 provides some example results of the correction criteria set.
[0113] Table 9. Examples of Head Impact Compensation Correction Basis
[0114] S41. Based on the head impact compensation correction basis, the master cylinder displacement compensation reference value in the head impact compensation reference is corrected to obtain the corrected master cylinder displacement compensation parameters.
[0115] In one feasible implementation, the main cylinder displacement compensation parameter is mainly used to characterize the impact of head stage reduction adjustment on head wall thickness control. Therefore, its correction is mainly affected by differences in continuous rolling load, differences in incoming material condition, differences in head wall thickness target, and differences in rolling speed. Among these, the impact of temperature condition differences on the main cylinder displacement compensation parameter is not directly incorporated into the main cylinder displacement compensation parameter correction rule in this embodiment, but is reflected in step S43 through the correction of the roll speed compensation parameter.
[0116] The corrected master cylinder displacement compensation parameters are defined as follows: (twenty three) In the formula, This indicates the corrected master cylinder displacement compensation parameters; a 1. a 2. a 3. a 4. a 5 indicates the correction coefficient for the master cylinder displacement compensation parameter; This indicates the threshold for segmentation of rolling speed differences.
[0117] In one feasible implementation, when the difference in continuous rolling load increases, it indicates that the deformation resistance of the target steel pipe head has changed relative to the reference state of similar historical batches, and the main cylinder displacement compensation parameter needs to be increased or decreased accordingly; when the difference in incoming material state increases, it indicates that the geometric state of the target steel pipe incoming material deviates from the reference state of similar historical batches, and the main cylinder displacement compensation parameter can be corrected according to preset rules; when the target head wall thickness changes, the main cylinder displacement compensation parameter can be incrementally corrected according to the target wall thickness control requirements; when the difference in rolling speed is small, the first stage speed correction coefficient is used for fine adjustment; when the difference in rolling speed exceeds the segment threshold, the second stage speed correction coefficient is used for larger-scale correction.
[0118] In this embodiment, taking target number T001 as an example, if the main cylinder displacement compensation reference value obtained in step S3 is 1.73mm, then... a 1 = 0.40 a 2 = 0.35 a 3 = 0.50 a 4 = 0.20 a 5 = 0.32 T spd =0.10, then the corrected main cylinder displacement compensation parameter can be obtained from equation (23) as approximately 1.78 mm.
[0119] S42. Based on the head impact compensation correction basis, the roll speed compensation reference value in the head impact compensation reference is corrected to obtain the corrected roll speed compensation parameters.
[0120] In one feasible implementation, the roll speed compensation parameter is mainly used to characterize the influence of changes in speed matching state, metal flow state and temperature state on head wall thickness control during the head stage. Therefore, its correction is mainly affected by differences in rolling speed, temperature state and continuous rolling load, while also taking into account differences in head wall thickness targets.
[0121] The corrected roll speed compensation parameters are defined as follows: (twenty four) In the formula, This indicates the corrected roll speed compensation parameters; b 1.b 2. b 3. b 4 represents the correction coefficient for the roll speed compensation parameter.
[0122] In one feasible implementation, when the rolling speed deviates from the reference state of similar historical batches, the roll speed compensation parameter can respond preferentially; when the temperature decreases or increases, the roll speed compensation parameter can be corrected accordingly based on the direction of the temperature difference; when the continuous rolling load increases, the metal flow state in the head stage can be improved by adjusting the roll speed compensation parameter; when the target head wall thickness changes, the roll speed compensation parameter can be adjusted in conjunction with the main cylinder displacement compensation parameter to form a correction result that matches the target wall thickness.
[0123] In this embodiment, taking target number T001 as an example, if the roll speed compensation reference value obtained in step S3 is 5.77 r / min, then... b 1 = 0.60 b 2 = -0.40 b 3 = 0.25 b 4=0.30, then the corrected roll speed compensation parameter can be obtained from equation (24) as approximately 5.79 r / min.
[0124] S43. Combining the coordinated matching relationship between the modified main cylinder displacement compensation parameters and the modified roll speed compensation parameters, as well as the target head wall thickness of the target steel pipe, the head impact compensation parameters of the target steel pipe are formed.
[0125] In one feasible implementation, the master cylinder displacement compensation parameter and the roll speed compensation parameter cannot be determined in isolation. This is because the master cylinder displacement compensation parameter primarily affects the reduction adjustment in the head stage, while the roll speed compensation parameter primarily affects the speed matching and metal flow state in the head stage; both together affect the head wall thickness control result. If only one parameter is modified, it can easily cause the other parameter to become mismatched with the current operating conditions, resulting in insufficient or excessive compensation.
[0126] In one feasible implementation, the cooperative matching relationship can be implemented using a rule table, empirical relationships, or joint correction rules. In this embodiment, a cooperative deviation determination method based on the consistency of the relative changes of two parameters is adopted, and the final head impact compensation parameters are formed when the cooperative deviation meets the requirements.
[0127] Define the two-parameter cooperative bias as: (25) In the formula, Ω represents the two-parameter cooperative deviation; C cyl This indicates the normalization scale for the master cylinder displacement compensation parameters.C spd This represents the normalization scale for roll speed compensation parameters. λ This represents the two-parameter synergy coefficient.
[0128] Furthermore, the impact compensation parameters for the target steel pipe head are defined as follows: (26) In the formula, This represents the head impact compensation parameter pair, which consists of the corrected master cylinder displacement compensation parameter and the corrected roll speed compensation parameter.
[0129] In one feasible implementation, when Ω does not exceed a preset coordination threshold T coop When, then take: (27) When Ω exceeds the preset coordination threshold T coop When that happens, the pre-set joint correction rules will be applied. and At least one of them is re-coordinated and corrected so that the corrected parameter combination meets the coordination threshold requirement, and finally the final head impact compensation parameters are obtained.
[0130] Table 10 provides an example of how the head impact compensation parameters for target number T001 are formed.
[0131] Table 10 Example of Head Impact Compensation Parameter Formation
[0132] Therefore, by extracting the difference information between the target steel pipe and similar historical batches, the main cylinder displacement compensation parameters and the roll speed compensation parameters are corrected respectively. Combined with the cooperative matching relationship between the two, a head impact compensation parameter that is compatible with the target steel pipe head wall thickness is formed, providing an input basis for subsequent equipment constraint verification and process constraint verification.
[0133] S5. Perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters to determine the target steel pipe head impact compensation parameters.
[0134] In one feasible implementation, the purpose of this step is to verify the head impact compensation parameters formed in step S4 with the equipment's executable boundaries and head wall thickness control requirements, and, if necessary, to perform constraint correction on parameters that exceed the constraint range, so as to obtain the final head impact compensation parameters that satisfy both the equipment implementation conditions and the head wall thickness control requirements.
[0135] Optionally, step S5 above may include the following steps S51-S52: S51. Extract the equipment constraint information and process constraint information corresponding to the continuous rolling control of the target steel pipe head. The equipment constraint information includes at least one or more of the following: allowable range of main cylinder displacement, allowable range of main cylinder adjustment rate, allowable range of roll speed, allowable range of speed change, and stand load limit. The process constraint information includes at least one or more of the following: head wall thickness control range, allowable head wall thickness deviation, and cooperative matching relationship of compensation parameters.
[0136] In one feasible implementation, the equipment constraint information may be derived from the rated parameters of the continuous rolling mill stand, the allowable stroke of the hydraulic actuator, the control system parameter table, the main drive speed regulation boundary, the upper limit of the stand design load, and the safety boundary formed by on-site commissioning; the process constraint information may be derived from product technical standards, order wall thickness requirements, process specifications, head wall thickness control rules, and the dual-parameter collaborative matching relationship determined in step S4.
[0137] For ease of representation, the set of device constraint information is defined as follows: (28) In the formula, Represents the set of equipment constraint information; and These represent the lower and upper limits of the allowable displacement of the master cylinder, respectively. Indicates the upper limit of the master cylinder adjustment rate; and These represent the lower and upper limits of the allowable roll speed, respectively. This indicates the upper limit of the allowable variation in roll speed; This indicates the maximum allowable load on the rack.
[0138] Furthermore, the set of process constraint information is defined as follows: (29) In the formula, Represents a set of process constraint information; T coop This represents the threshold for collaborative matching of compensation parameters.
[0139] In one feasible implementation, the head wall thickness control range can be directly calculated from the head wall thickness target value and allowable deviation in step S2, or the upper and lower limit range can be directly given by the process specification; the compensation parameter coordination matching relationship can be represented by the two-parameter coordination deviation and coordination threshold in step S4.
[0140] Table 11 provides some examples of equipment constraint information and process constraint information.
[0141] Table 11 Examples of Equipment Constraint Information and Process Constraint Information
[0142] S52. The corrected main cylinder displacement compensation parameters and the corrected roll speed compensation parameters are compared with the equipment constraint information and the process constraint information, respectively. The compensation parameters that exceed the constraint range are corrected within the constraint range. The feasibility and consistency of the head wall thickness target are judged for the corrected candidate head impact compensation parameters. The final head impact compensation parameters of the target steel pipe are determined.
[0143] In one feasible implementation, the head impact compensation parameters formed in step S4 are used. As input, the main cylinder displacement compensation parameters and roll speed compensation parameters are first compared with the allowable range of main cylinder displacement and roll speed in the equipment constraint information; when any compensation parameter exceeds the corresponding allowable range, the constraint correction is performed by boundary truncation.
[0144] The candidate values for the main cylinder displacement after boundary truncation can be expressed as: (30) In the formula, This represents the candidate values for the main cylinder displacement after boundary truncation; This represents the candidate value for the roll speed after boundary truncation.
[0145] Furthermore, the set of candidate head impact compensation parameters is defined as follows: (31) In the formula, This represents a set of candidate head impact compensation parameters.
[0146] In one feasible implementation, "feasibility" means that the candidate head impact compensation parameters simultaneously meet the execution boundaries defined by the equipment constraint information, including the allowable range of main cylinder displacement, the allowable range of main cylinder adjustment rate, the allowable range of roll speed, the allowable range of speed change, and the allowable upper limit of frame load; the head wall thickness target consistency means that under the action of the candidate head impact compensation parameters, the predicted result of the target steel pipe head wall thickness falls within the head wall thickness control range and meets the head wall thickness allowable deviation requirements.
[0147] In one feasible implementation, the predicted head wall thickness corresponding to the candidate head impact compensation parameters can be estimated using a local linear approximation method, expressed as: (32) In the formula, This indicates the predicted head wall thickness. This represents the reference header wall thickness value corresponding to a set of similar historical batches; k 1 and k2 represents the local sensitivity coefficients of the influence of changes in main cylinder displacement and roll speed on the head wall thickness, respectively.
[0148] Candidate head impact compensation parameters are considered to meet the head wall thickness target consistency requirement when the following conditions are met: (33) In one feasible implementation, this embodiment uses a single candidate parameter set to directly determine the final parameters. That is, when the candidate head impact compensation parameter set... When both feasibility and head wall thickness target requirements are met, the parameter is directly determined as the final head impact compensation parameter for the target steel pipe.
[0149] Therefore, the final head impact compensation parameter of the target steel pipe is defined as follows: (34) In the formula, This indicates the final head impact compensation parameters for the target steel pipe. This indicates the final master cylinder displacement compensation parameters. This indicates the final roll speed compensation parameters.
[0150] When the candidate head impact compensation parameter set When both feasibility and head wall thickness target requirements are met, the following can be chosen: (35) When the candidate head impact compensation parameter set If the feasibility requirement is met but the head wall thickness target consistency requirement is not met, the main cylinder displacement compensation parameter and the roll speed compensation parameter are further coordinated and corrected in the neighborhood of the candidate head impact compensation parameter to generate a set of corrected candidate parameters. The set that minimizes the head wall thickness prediction deviation and still meets the equipment constraints and process constraints is selected as the final head impact compensation parameter.
[0151] In one feasible implementation, the further coordination correction can be achieved by a graded increase or decrease method, that is, within a preset correction step size range, the main cylinder displacement compensation parameter and the roll speed compensation parameter are adjusted by increment or decrease until the head wall thickness prediction value falls within the head wall thickness control range.
[0152] Table 12 provides examples of equipment constraint verification and process constraint verification for target number T001.
[0153] Table 12 Examples of Equipment Constraint Verification and Process Constraint Verification
[0154] Therefore, by extracting the equipment constraint information and process constraint information corresponding to the continuous rolling control of the target steel pipe head, and comparing, correcting within constraints, and judging the feasibility and consistency of the target head wall thickness with the corrected compensation parameters formed in step S4, the final head impact compensation parameters of the target steel pipe are determined, providing an input basis for subsequent execution feedback and knowledge base updates.
[0155] S6. Collect the head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, update the head impact compensation knowledge base according to the process feedback data and the head wall thickness feedback data, and correct the determination results of the head impact compensation parameters of subsequent steel pipes.
[0156] In one feasible implementation, the purpose of this step is to collect head process feedback data and head wall thickness feedback data after the target steel pipe completes the head continuous rolling, calculate the head wall thickness control deviation index, and write the target head working condition characteristics, final head impact compensation parameters, feedback data and deviation index as new knowledge entries into the head impact compensation knowledge base; at the same time, when the deviation exceeds the preset range, the compensation benchmark selection basis, parameter correction rules or constraint correction basis used to determine the subsequent steel pipe head impact compensation parameters are corrected.
[0157] Optionally, step S6 above may include the following steps S61-S63: S61. Collect head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, and calculate the head wall thickness control deviation index based on the head wall thickness feedback data, the target head wall thickness control requirements, and the final head impact compensation parameters corresponding to the target steel pipe.
[0158] In one feasible implementation, the head process feedback data includes at least the stand load feedback data, rolling speed feedback data, main cylinder displacement feedback data, and roll speed feedback data within the head control interval; the head wall thickness feedback data includes at least the measured wall thickness sequence within the head control interval. The sampling position of the feedback data corresponds to the head control length interval of the steel pipe, and the sampling time range corresponds to the time range from the steel pipe head entering the continuous rolling mill stand to the end of the head control interval.
[0159] Define the head wall thickness deviation sequence as follows: (36) (37) In the formula, This represents the head process feedback data set. This represents the head wall thickness feedback data set. , , and They represent the first n Feedback values of stand load, rolling speed, main cylinder displacement, and roll speed corresponding to each sampling point. Indicates the first n The measured wall thickness value corresponding to each sampling point N This indicates the number of sampling points within the head control zone.
[0160] Define the head wall thickness deviation sequence as follows: (38) In the formula, Δ w n Indicates the first n The head wall thickness deviation corresponding to each sampling point.
[0161] Furthermore, the head wall thickness control deviation index can be expressed using the average deviation, maximum absolute deviation, and out-of-tolerance ratio: (39) (40) (41) In the formula, Indicates the average deviation of the head wall thickness, Δ w peak This indicates the maximum absolute deviation of the head wall thickness. η w This indicates the percentage of head wall thickness exceeding tolerance.
[0162] Table 13 provides some examples of head feedback data and deviation indicators.
[0163] Table 13 Examples of Head Feedback Data and Deviation Indicators
[0164] S62. The target head working condition characteristics, final head impact compensation parameters, head process feedback data, head wall thickness feedback data and head wall thickness control deviation index of the target steel pipe are associated and stored to form an updated head impact compensation knowledge entry, and the head impact compensation knowledge base is updated.
[0165] In one feasible implementation, after obtaining the head process feedback data, head wall thickness feedback data, and head wall thickness control deviation index, the data is associated with the target head working condition characteristics and the final head impact compensation parameters of the target steel pipe to form an updated head impact compensation knowledge entry.
[0166] The updated head impact compensation knowledge entry is defined as follows: (42) In the formula, This indicates the updated head impact compensation knowledge entry.
[0167] Furthermore, the updated head impact compensation knowledge base can be represented as: (43) In the formula, This refers to the updated head impact compensation knowledge base.
[0168] In one feasible implementation, in this embodiment, the head impact compensation knowledge base is updated by writing a knowledge entry after each steel pipe completes the head continuous rolling; in another implementation, it can also be updated uniformly after batch aggregation or according to a preset cycle.
[0169] S63. Based on the head wall thickness control deviation index and the head process feedback data, correct the compensation benchmark selection basis, parameter correction rules, or constraint correction basis used when determining the subsequent steel pipe head impact compensation parameters.
[0170] In one feasible implementation, when the head wall thickness control deviation index is within the allowable range, only the knowledge entry is entered into the database; when the head wall thickness control deviation index exceeds a preset threshold, subsequent steel pipe parameter determination rule correction is triggered. The rule correction includes at least one or more of the following: compensation benchmark selection basis correction, parameter correction rule correction, and constraint correction basis correction.
[0171] Define the update trigger indicator as follows: (44) In the formula, ψ This indicates the update trigger indicator quantity. This indicates the trigger threshold for the average deviation of the head wall thickness. This indicates the trigger threshold for the maximum absolute deviation of the head wall thickness. T η This indicates the threshold for triggering the error in the proportion of head wall thickness exceeding tolerance.
[0172] In one feasible implementation, when ψ When =1, at least one of the following correction methods can be performed: 1. Revise the selection criteria for compensation benchmarks. That is, update the comprehensive weight of each historical knowledge item in step S33 based on the head wall thickness control effect of each historical knowledge item.
[0173] (45)
[0174] In the formula, For the first u The overall weight of each updated historical knowledge item; ρu The overall weights before the update; , and For the first u The average deviation, maximum absolute deviation, and out-of-tolerance ratio of the header wall thickness of each historical knowledge item; 1. 2 and 3 represents the preset evaluation coefficient.
[0175] The worse the control effect, the smaller the exponent and the lower the corresponding overall weight; the better the control effect, the higher the corresponding overall weight. In subsequent step S33, the updated [explanation / revision] will be used. The main cylinder displacement compensation benchmark value and the roll speed compensation benchmark value are recalculated using weighted recalculation.
[0176] 2. Modify the parameter correction rules, that is, when the actual execution deviation of the main cylinder displacement and the roll speed is within the allowable range, update the main cylinder displacement correction coefficient in step S42, the roll speed correction coefficient in step S43, or the rolling speed difference segment threshold according to the average deviation of the head wall thickness of the target steel pipe and the working condition difference obtained in step S41.
[0177] In one feasible implementation, correction input vectors corresponding to the main cylinder displacement compensation parameters and the roll speed compensation parameters are constructed respectively: (46) In the formula, cyl The main cylinder displacement correction input vector; spd The input vector for roll speed correction; 1(·) is the indicator function; Δ (1) ~Δ (5) These are the differences in continuous rolling load, rolling speed, temperature conditions, incoming material conditions, and target head wall thickness obtained in step S41.
[0178] Based on the average deviation of the head wall thickness, the correction coefficients for the main cylinder displacement and the roll speed have been updated: (47) In the formula, and These are the updated master cylinder displacement correction coefficient and roll speed correction coefficient, respectively. μ a and μ b Preset update rate; k cyl and k spdThese are the local sensitivity of the main cylinder displacement compensation parameters and the roll speed compensation parameters to the head wall thickness, respectively. This represents the average deviation of the head wall thickness in this study; clip( This indicates that the updated correction factor will be limited to the preset upper and lower limits.
[0179] When the steel pipe is subsequently processed in steps S42 and S43, it is as follows: Replacement of the original master cylinder displacement correction coefficient a i ,by Replacement of original roll speed correction factor b j And recalculate the corresponding main cylinder displacement compensation correction amount and roll speed compensation correction amount.
[0180] If it is necessary to adjust the segmented threshold for rolling speed difference, then select the threshold from the preset candidate threshold set that minimizes the prediction error of the head wall thickness of the recently fed-in steel pipe: (48) In the formula, The updated threshold for segmentation of rolling speed differences; For a preset set of candidate thresholds; M This refers to the recent feedback on the quantity of steel pipes; To use candidate thresholds T The calculated first m Predicted wall thickness at the head of the branch steel pipe; This corresponds to the actual head wall thickness.
[0181] 3. The constraint correction basis is modified. That is, when the number of steel pipes with consecutive head wall thickness deviations in the same direction reaches the preset trigger number, the search step size, local search range and dual-parameter collaborative matching threshold of the subsequent candidate compensation parameters are adjusted.
[0182] First, adjust the step size for correcting the candidate parameters: (49) In the formula, Adjust the step size for the updated candidate parameters; h q Adjust the step size for the original candidate parameters; This is the maximum permissible correction step size; N dir The number of steel pipes that exhibit continuous wall thickness deviations in the same direction; N trig The preset trigger quantity; μ h This is the step size adjustment factor.
[0183] Based on the updated correction step size, construct the local search range for subsequent candidate parameters: (50) In the formula, This is the updated local search range; The final compensation parameters used for the target steel pipe in this study; and These represent the lower and upper limits allowed for the corresponding devices.
[0184] Simultaneously tighten the dual-parameter collaborative matching threshold: (51) In the formula, The updated two-parameter collaborative matching threshold; T coop This is the original collaborative matching threshold; This is the lower limit of the collaborative matching threshold; μ T This is the threshold adjustment coefficient.
[0185] Subsequent steel pipes in the updated local search range Within, according to the updated correction step size Generate candidate compensation parameters and apply the updated collaborative matching threshold. The screening process continues. The equipment's rated allowable range remains a hard constraint and does not change based on feedback results.
[0186] In one feasible implementation, this embodiment employs a method of writing each steel pipe into a knowledge base and triggering rule correction when the deviation exceeds a threshold; when ψ When =0, only the knowledge entry is added to the database, and no subsequent rule modification is triggered.
[0187] Table 14 provides some examples of rule correction triggers.
[0188] Table 14 Example of rule correction triggering
[0189] Therefore, by collecting feedback data on the head process and head wall thickness after continuous rolling of the target steel pipe, the head wall thickness control deviation index is calculated, and the target head working condition characteristics, final head impact compensation parameters, feedback data and deviation index are written into the head impact compensation knowledge base. At the same time, when the deviation exceeds the threshold, the rules used to determine the subsequent steel pipe head impact compensation parameters are corrected, thereby realizing the rolling update of the head impact compensation knowledge base and the continuous correction of the subsequent parameter determination rules.
[0190] Figure 5This is a schematic diagram of a device and electronic equipment for determining impact compensation parameters for the head wall thickness control of seamless steel pipe continuous rolling mill, provided in an embodiment of the present invention. Figure 5 As shown, the seamless steel pipe continuous rolling head wall thickness control impact compensation parameter determination device may include a knowledge base construction unit 510, a working condition characterization unit 520, a benchmark determination unit 530, a parameter correction unit 540, a parameter determination unit 550, and a feedback update unit 560.
[0191] The knowledge base construction unit 510 is used to collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data and historical compensation parameter records related to the continuous rolling control of the steel pipe head in historical stable production batches, and to perform structured processing on the data to construct a head impact compensation knowledge base.
[0192] The working condition characterization unit 520 is used to acquire the specification information, head wall thickness control target and current continuous rolling working condition information of the target steel pipe before the target steel pipe enters the continuous rolling head control stage, and to construct the target head working condition characteristics.
[0193] The benchmark determination unit 530 is used to determine a set of similar historical batches in the head impact compensation knowledge base according to the working condition characteristics of the target head, and to determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches.
[0194] The parameter correction unit 540 is used to correct the head impact compensation benchmark by combining the current continuous rolling load, rolling speed, temperature state, incoming material state and head wall thickness target of the target steel pipe, so as to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe.
[0195] The parameter determination unit 550 is used to perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters, and to determine the target steel pipe head impact compensation parameters.
[0196] The feedback update unit 560 is used to collect head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, update the head impact compensation knowledge base according to the process feedback data and the head wall thickness feedback data, and correct the determination results of the head impact compensation parameters of subsequent steel pipes.
[0197] Furthermore, such as Figure 5As shown, the electronic device may include a processor, a memory, and a communication interface. The memory stores computer-readable instructions, and the processor calls and executes these instructions to implement the steps in the method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes. The communication interface interacts with the production process data acquisition system, the head wall thickness detection system, the continuous rolling control system, and the historical knowledge base. The electronic device may be an edge computing node of the continuous rolling production line, a factory server, or a cloud computing platform.
[0198] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for determining impact compensation parameters for head wall thickness control in continuous rolling of seamless steel pipes, characterized in that, The method includes: S1. Preset the head control range, collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data and historical compensation parameter records related to the continuous rolling control of the steel pipe head from historical stable production batches, and perform structured processing on the collected information and data to build a head impact compensation knowledge base. S2. Before the target steel pipe enters the continuous rolling head control stage, acquire the target steel pipe's specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control target, current continuous rolling working condition information, and head control interval information, and construct the target head working condition characteristics. S3. Based on the target head working condition characteristics, determine a set of similar historical batches in the head impact compensation knowledge base, and determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches; wherein, the head impact compensation benchmark includes the main cylinder displacement compensation benchmark value and the roll speed compensation benchmark value; S4. Based on the current continuous rolling load, rolling speed, temperature status, incoming material status and head wall thickness target of the target steel pipe, the head impact compensation benchmark is corrected to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe. S5. Perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters to determine the target steel pipe head impact compensation parameters. S6. Collect the head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, and update the head impact compensation knowledge base according to the head process feedback data and head wall thickness feedback data to correct the determination results of the head impact compensation parameters of subsequent steel pipes.
2. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, The historical compensation parameter records in S1 include historical master cylinder displacement compensation parameters and roll speed compensation parameter records.
3. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, The process of constructing the head impact compensation knowledge base in S1 is as follows: S11. Based on the timing information of the steel pipe head entering the continuous rolling mill stand and the preset head control interval, extract the head stage data related to head impact compensation from each historical stable production batch, and associate the head stage data, head wall thickness detection data and historical compensation parameter records with the steel pipe number, stand number and time sequence to form head impact compensation knowledge entries. The head stage data includes rolling force, rolling speed, main cylinder displacement response, roll speed response, stand load, temperature status, and tension status for each stand. S12. The head impact compensation knowledge entries are structured and stored to obtain a head impact compensation knowledge base: S121. Using the steel pipe number as the main association identifier of the head impact compensation knowledge entry, and the frame number and sampling time as the sub-association identifiers, the data corresponding to the same steel pipe are divided into the basic attribute field group, the head stage process field group, the compensation parameter field group and the wall thickness result field group, and written into the same head impact compensation knowledge entry. The basic attribute field group includes specification category, steel grade category, target wall thickness, and head control range; the head stage process field group includes rolling force, rolling speed, master cylinder displacement response, roll speed response, stand load, temperature status, and tension status for each stand; the compensation parameter field group includes historical master cylinder displacement compensation parameters and historical roll speed compensation parameters; and the wall thickness result field group includes average head wall thickness, maximum absolute head deviation, and head wall thickness deviation ratio. S122. Generate specification category label, steel grade category label, wall thickness target category label and continuous rolling condition category label according to the specification category, steel grade category, wall thickness target category label and continuous rolling condition category label in the head impact compensation knowledge entry, and form a composite retrieval index with the four category labels. Use the composite retrieval index to store and retrieve each head impact compensation knowledge entry to obtain the head impact compensation knowledge base. The continuous rolling condition category is a condition category label generated based on the head stage process field group, including high load type, medium load type, high speed type, low temperature type, or composite condition type.
4. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, The specifications in S2 include outer diameter, wall thickness and length; the incoming material parameters include incoming material size and incoming material temperature; and the current continuous rolling condition information includes stand load, rolling speed, rotation speed status and temperature status. The specification parameters, steel grade parameters, incoming material parameters, continuous rolling setting parameters, head wall thickness control target, current continuous rolling condition information, and head control range information are combined and characterized to form the target head condition characteristics.
5. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, Specifically, S3 refers to: S31. Based on the target head working condition characteristics, extract candidate historical batches that match the target steel pipe in terms of specification category, steel type category, wall thickness target category and continuous rolling working condition category from the head impact compensation knowledge base. S32. A comprehensive evaluation is conducted on the differences in specifications, incoming materials, settings, working conditions, and target head wall thickness between the candidate historical batches and the target steel pipe. Based on the similarity evaluation value obtained from the comprehensive evaluation and the head wall thickness compliance results corresponding to each candidate historical batch, the candidate historical batches are screened and sorted to determine the set of similar historical batches. The comprehensive evaluation is as follows: calculate the normalized difference between the candidate historical batches and the target steel pipe in each comparison field, and perform a weighted summation of each normalized difference to obtain the similarity evaluation value corresponding to each candidate historical batch. S33. Extract the historical compensation parameters corresponding to the set of similar historical batches, and determine the target steel pipe head impact compensation benchmark by combining the head wall thickness results of each similar historical batch.
6. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, Specifically, S4 refers to: S41. Based on the differences in continuous rolling load, incoming material condition, target head wall thickness, and rolling speed, as well as the correction coefficient of the main cylinder displacement compensation parameter, calculate the correction amount of the main cylinder displacement compensation. By adding the main cylinder displacement compensation reference value to the main cylinder displacement compensation correction amount, the corrected main cylinder displacement compensation parameter is obtained. S42. Based on the differences in rolling speed, temperature conditions, continuous rolling load, and target head wall thickness, as well as the correction coefficient of the roll speed compensation parameter, calculate the roll speed compensation correction amount. By adding the roll speed compensation reference value to the roll speed compensation correction amount, the corrected roll speed compensation parameter is obtained. S43. Combining the coordinated matching relationship between the modified main cylinder displacement compensation parameters and the modified roll speed compensation parameters, and the target head wall thickness of the target steel pipe, the head impact compensation parameters of the target steel pipe are formed; wherein, the head impact compensation parameters of the target steel pipe include the main cylinder displacement compensation parameters and the roll speed compensation parameters.
7. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, Specifically, S5 is: S51. Extract the equipment constraint information and process constraint information corresponding to the continuous rolling control of the target steel pipe head; The equipment constraint information includes the allowable range of main cylinder displacement, the allowable range of main cylinder adjustment rate, the allowable range of roll speed, the allowable range of speed variation, and the frame load limit. The process constraint information includes the head wall thickness control range, the head wall thickness allowable deviation, and the compensation parameter collaborative matching threshold. S52. Compare the main cylinder displacement compensation parameters and roll speed compensation parameters obtained in S4 with the equipment constraint information and process constraint information in S51 respectively. Correct the compensation parameters that exceed the constraint range within the constraint range. Then, judge the feasibility and consistency of the head wall thickness target for the corrected candidate head impact compensation parameters to determine the final head impact compensation parameters of the target steel pipe.
8. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, The update process in S6 is as follows: S61. Based on the difference between the measured wall thickness value of each sampling point in the head control interval and the target value of the head wall thickness, calculate the average deviation of the head wall thickness, the maximum absolute deviation of the head wall thickness, and the proportion of head wall thickness exceeding the tolerance, which are used as head wall thickness control deviation indicators. S62. Write the target head working condition characteristics, final head impact compensation parameters, head process feedback data, head wall thickness feedback data and head wall thickness control deviation index into the same head impact compensation knowledge entry to form an updated head impact compensation knowledge entry and update the head impact compensation knowledge base. S63. Based on the head wall thickness control deviation index and the head process feedback data, when any one of the head wall thickness control deviation indices exceeds the corresponding preset threshold, a correction is triggered; the selection basis for the compensation benchmark, parameter correction rules, or constraint correction basis used when determining the subsequent steel pipe head impact compensation parameters are corrected; if the threshold is not exceeded, only the production data of this production is written to the knowledge base.
9. The method for determining impact compensation parameters for head wall thickness control in continuous rolling mills of seamless steel pipes according to claim 1, characterized in that, The apparatus for implementing the method includes: The knowledge base construction unit is used to collect specification information, steel grade information, process setting data, process operation data, head wall thickness detection data, and historical compensation parameter records related to the continuous rolling control of the steel pipe head from historical stable production batches, and to perform structured processing on the data to construct a head impact compensation knowledge base; The working condition characterization unit is used to acquire the specification information, head wall thickness control target and current continuous rolling working condition information of the target steel pipe before the target steel pipe enters the continuous rolling head control stage, and to construct the target head working condition characteristics. The benchmark determination unit is used to determine a set of similar historical batches in the head impact compensation knowledge base according to the working condition characteristics of the target head, and to determine the target steel pipe head impact compensation benchmark based on the set of similar historical batches. The parameter correction unit is used to correct the head impact compensation benchmark by combining the current continuous rolling load, rolling speed, temperature state, incoming material state and head wall thickness target of the target steel pipe, so as to obtain the main cylinder displacement compensation parameters and roll speed compensation parameters of the target steel pipe. The parameter determination unit is used to perform equipment constraint verification and process constraint verification on the main cylinder displacement compensation parameters and the roll speed compensation parameters, and to determine the target steel pipe head impact compensation parameters. The feedback update unit is used to collect head process feedback data and head wall thickness feedback data after continuous rolling of the target steel pipe, update the head impact compensation knowledge base according to the process feedback data and the head wall thickness feedback data, and correct the determination results of head impact compensation parameters for subsequent steel pipes.
