A method for predicting and self-correcting the size deviation of a steel pipe during on-line sizing process

Abstract

This invention relates to online sizing control technology for steel pipes, and more particularly to a method for predicting and self-correcting dimensional deviations during the online sizing process of steel pipes. The method divides the steel pipe into continuous segments along its length, establishes a segment state vector, inputs a dimensional deviation prediction model, calculates the predicted outer diameter deviation and predicted ellipticity deviation for each segment, and obtains the deviation source weights corresponding to temperature changes, wall thickness changes, tension changes, roll gap offset, and roll wear. Before the segment reaches the adjustable sizing stand, the predicted deviation and deviation source weights are input into the self-correcting controller to generate roll gap compensation and tension compensation amounts, which are then applied to the target sizing stand in advance based on the current position and running speed of the steel pipe. The method also calculates the predicted residual using the post-diameter measurement results and updates the deviation source weights and control gain parameters of the self-correcting controller based on the residual, achieving segment-level predictive control and closed-loop self-correction.

Description

Technical Field

[0001] This invention belongs to the field of steel pipe rolling and finishing control technology, and in particular relates to a method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes. Background Technology

[0002] Online sizing of steel pipes is a crucial finishing process after pipe forming. It is typically performed on a continuous production line following rolling, cooling, or heat treatment. Multi-stand sizing rollers are used to press and shape the outer diameter of the steel pipe, ensuring it meets specified geometric dimensional requirements. In current production, the sizing process relies heavily on parameters such as the roller gap setting of the sizing mill, rolling process parameters, incoming material temperature, pipe wall thickness, steel grade characteristics, and rolling speed. Operators determine the initial roller gap based on specifications, experience models, or process databases. Diameter gauges, temperature detection devices, speed detection devices, and PLC control systems on the production line collect data on the pipe's operating status. Some production lines also incorporate laser diameter measurement, infrared thermography, online data recording, and automatic alarm functions to determine if the pipe's outer diameter is within acceptable limits.

[0003] The core problem with existing online sizing control for steel pipes lies in the fact that dimensional deviation control still relies primarily on corrections based on the test results, lacking predictive control and self-correcting closed-loop systems built around the controlled non-electrical variable of dimensional deviation. The outer diameter or ellipticity data obtained by online sizing equipment are mostly results after the steel pipe has undergone sizing. These results only indicate that dimensional deviation has occurred, but cannot predict the trend of deviation formation under the combined influence of changes in incoming material temperature, wall thickness fluctuations, tension changes, roll gap drift, and roll system wear. After receiving signals of exceeding tolerances or increasing deviation, the control system typically adjusts control quantities such as roll gap, speed, and tension manually based on experience or fixed rules. There is a lag between the adjustment action and the formation of the deviation, and the adjustment range is difficult to automatically change with the deviation trend. Summary of the Invention

[0004] The purpose of this invention is to provide a method for predicting and self-correcting deviations in the online sizing process of steel pipes. By combining the pipe segment state vector with the dimensional deviation prediction model and applying the self-correcting controller in advance, this invention solves the problems in the prior art that rely solely on post-diameter measurement results for correction, control lag, and the inability of the roll gap and tension adjustment amplitude to automatically change with the deviation trend.

[0005] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution.

[0006] This invention is a method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes, comprising: S1. acquiring the incoming material status parameters, the execution parameters of the sizing machine stand and the online detection parameters during the online sizing process of the steel pipe; dividing the steel pipe into continuous pipe segments along the length direction according to the running speed of the steel pipe, the sampling time and the position of each sizing machine stand, and establishing the pipe segment status vector of each pipe segment; S2. Input the pipe segment state vector into the size deviation prediction model, calculate the predicted outer diameter deviation and predicted ellipticity deviation of the corresponding pipe segment before reaching the post-diameter measurement position, and obtain the deviation source weights corresponding to temperature change, wall thickness change, tension change, roll gap offset and roll system wear. S3. Input the predicted outer diameter deviation, predicted ellipticity deviation and deviation source weight into the self-correcting controller, and generate the roll gap compensation amount and tension compensation amount to offset the predicted outer diameter deviation and predicted ellipticity deviation before the corresponding pipe section reaches the adjustable sizing frame. S4. Based on the running speed of the steel pipe, the current position of the pipe section, and the position of the sizing machine stand, the roll gap compensation amount and tension compensation amount are matched in time and applied when the corresponding pipe section reaches the target sizing machine stand; S5. Obtain the measured outer diameter and measured ellipticity of the corresponding pipe segment after leaving the sizing unit. Compare the measured outer diameter and measured ellipticity with the predicted outer diameter deviation and predicted ellipticity deviation, respectively, to obtain the predicted residual. Update the deviation source weight and the control gain parameter of the self-correcting controller based on the predicted residual for subsequent pipe segment size deviation prediction and self-correcting control.

[0007] The present invention is further configured such that the incoming material status parameters include the target outer diameter of the steel pipe, the target ellipticity, the allowable deviation of the outer diameter, the allowable deviation of the ellipticity, the wall thickness of the steel pipe, the steel grade of the steel pipe, the temperature of the steel pipe before entering the sizing unit, the running speed of the steel pipe, and the outer diameter detection value of the steel pipe before entering the sizing unit. The online detection parameters include the measured outer diameter, the measured ellipticity, the circumferential distribution value of the outer diameter, and the pipe section detection time collected by the post-diameter measuring device set on the outlet side of the sizing unit. The measured outer diameter and the measured ellipticity are used to form the feedback dimension data of the corresponding pipe section.

[0008] The present invention is further configured such that when the steel pipe described in S1 is divided into continuous pipe segments along the length direction, the length of a single pipe segment is determined by the sampling period and the running speed of the steel pipe, and a pipe segment number is assigned to each pipe segment so that the same pipe segment maintains the same pipe segment number when entering the sizing unit, passing through each sizing frame and reaching the post-diameter measuring position.

[0009] The present invention is further configured such that the pipe segment state vector is represented as: X i =[ΔD i0 ΔE i0 ΔTi ΔH i , ΔV i , ΔF ij ΔG ij ΔR ij W ij ], Where i represents the pipe section number, j represents the sizing stand number, and ΔD i0 Indicates the deviation of the inlet outer diameter, ΔE i0 Indicates the inlet ellipticity deviation, ΔT i ΔH represents the change in temperature. i ΔV represents the change in wall thickness. i ΔF represents the velocity fluctuation. ij ΔG represents the change in tension. ij Indicates the roll gap offset, ΔR ij W represents the rolling force deviation. ij This indicates the amount of wear on the roller system.

[0010] The present invention is further configured such that the roll gap offset ΔG ij Determined according to the following relationship: ΔG ij =G ij f -G ij s -C ij T , Among them, G ij f G represents the roll gap feedback value corresponding to the j-th sizing stand for the i-th pipe segment. ij s C represents the roll gap setting value for the j-th sizing stand. ij T This indicates the amount of heat compensation for the roll gap caused by the temperature rise of the machine frame.

[0011] The present invention is further configured such that the dimensional deviation prediction model includes an outer diameter deviation prediction channel and an ellipticity deviation prediction channel; the outer diameter deviation prediction channel is used to calculate and predict the outer diameter deviation based on the temperature change, wall thickness change, tension change, roll gap offset, and roll system wear; the ellipticity deviation prediction channel is used to calculate and predict the ellipticity deviation based on the temperature change, wall thickness change, tension change, roll gap offset, and roll system wear.

[0012] The present invention is further configured such that the deviation source weight includes a weight magnitude and an action direction, wherein the weight magnitude is used to represent the degree of contribution of the corresponding deviation source to the predicted outer diameter deviation or the predicted ellipticity deviation, and the action direction is used to represent whether the corresponding deviation source increases or decreases the outer diameter deviation or the ellipticity deviation.

[0013] The present invention is further configured such that the self-correcting controller includes a deviation judgment unit, a compensation amount generation unit, and a control amount allocation unit. The deviation judgment unit is used to determine whether the predicted outer diameter deviation and the predicted ellipticity deviation exceed a preset deviation threshold. The compensation amount generation unit is used to generate a roll gap compensation amount and a tension compensation amount. The control amount allocation unit is used to determine the allocation ratio of the roll gap compensation amount and the tension compensation amount. When the deviation source weight corresponding to roll gap offset or roll system wear is greater than a preset equipment disturbance weight threshold, the control amount allocation unit increases the allocation ratio of the roll gap compensation amount. When the deviation source weight corresponding to tension change is greater than a preset tension disturbance weight threshold, the control amount allocation unit increases the allocation ratio of the tension compensation amount.

[0014] The present invention is further configured such that the roll gap compensation amount is constrained by the maximum adjustment amount of the roll gap in a single operation, the maximum adjustment difference of the roll gap between adjacent pipe sections, and the response time of the roll gap actuator; the tension compensation amount is constrained by the maximum adjustment amount of the tension in a single operation, the maximum adjustment difference of the tension between adjacent pipe sections, and the response time of the tension adjustment mechanism; when the difference in the roll gap compensation amount or the difference in the tension compensation amount between adjacent pipe sections exceeds the corresponding maximum adjustment difference, the roll gap compensation amount or the tension compensation amount is subjected to smoothing and limiting processing, so that the control amount between continuous pipe sections increases or decreases according to a preset rate of change.

[0015] The present invention is further configured such that, in step S4, the timing matching determines the control command output time based on the current position of the corresponding pipe segment, the position of the target sizing machine, the running speed of the steel pipe, and the response time of the actuator, and the control command output time is earlier than the time when the corresponding pipe segment arrives at the target sizing machine.

[0016] The beneficial effects of the present invention are as follows: The present invention divides the steel pipe into continuous pipe segments and establishes a pipe segment state vector. It inputs the size deviation prediction model to calculate the predicted outer diameter deviation and the predicted ellipticity deviation, thereby realizing the early prediction at the pipe segment level and avoiding the passive correction that relies solely on the results of subsequent diameter measurement in the traditional method.

[0017] The self-correcting controller generates roll gap and tension compensation amounts based on the predicted deviation and the weight of the deviation source, and performs timing matching to apply them in advance to the target sizing frame, thereby achieving active closed-loop control and reducing control lag and adjustment overshoot.

[0018] The system calculates the predicted residuals by using the post-diameter measurement results and updates the deviation source weights and control gain parameters in real time, enabling the system to adaptively adjust the compensation strategy for different pipe sections and operating conditions, thereby improving the control accuracy and consistency of the steel pipe's outer diameter and ellipticity.

[0019] This method comprehensively considers factors such as temperature changes, wall thickness fluctuations, tension changes, roll gap offset, and roll system wear, and achieves targeted control for different pipe sections, significantly improving the dimensional fluctuations of the head section, stable section, and tail section, and ensuring high quality and high efficiency in the steel pipe production process. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0021] Figure 1 This is a flowchart illustrating a method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes.

[0022] Figure 2 This is a schematic diagram of an online sizing system for steel pipes, which is a method for predicting dimensional deviations and controlling self-correction during the online sizing process.

[0023] Figure 3 This diagram illustrates the self-correcting controller and roll gap offset calculation method for a dimensional deviation prediction and self-correcting control method in the online sizing process of steel pipes. Detailed Implementation

[0024] The technical solutions of the present invention will be described below with reference to the accompanying drawings. The described embodiments are only some embodiments of the present invention, and not all embodiments.

[0025] Example 1

[0026] Please see Figures 1-3 This invention relates to a method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes. The method includes: S1. Obtaining incoming material state parameters, sizing mill execution parameters, and online detection parameters during the online sizing process; dividing the steel pipe into continuous segments along its length based on the pipe's running speed, sampling time, and the position of each sizing mill; and establishing a segment state vector for each segment. The incoming material state parameters include the target outer diameter of the steel pipe, target ellipticity, allowable deviation of the outer diameter, allowable deviation of the ellipticity, steel pipe wall thickness, steel pipe grade, temperature of the steel pipe before entering the sizing mill, and the steel pipe running speed. The online detection parameters include the measured outer diameter, measured ellipticity, circumferential distribution value of outer diameter, and pipe section detection time collected by the post-diameter measuring device set on the outlet side of the sizing unit. The measured outer diameter and measured ellipticity are used to form the feedback dimension data of the corresponding pipe section. When the steel pipe is divided into continuous pipe sections along the length direction, the length of a single pipe section is determined by the sampling period and the running speed of the steel pipe, and a pipe section number is assigned to each pipe section so that the same pipe section maintains the same pipe section number when entering the sizing unit, passing through each sizing frame, and reaching the post-diameter measuring position.

[0027] S2. Input the pipe segment state vector into the dimensional deviation prediction model, calculate the predicted outer diameter deviation and predicted ellipticity deviation of the corresponding pipe segment before reaching the post-diameter measurement position, and obtain the deviation source weights corresponding to temperature change, wall thickness change, tension change, roll gap offset, and roll system wear. The pipe segment state vector is represented as: X i =[ΔD i0 ΔE i0 ΔT i ΔH i , ΔV i , ΔF ij ΔG ij ΔR ij W ij ], Where i represents the pipe section number, j represents the sizing stand number, and ΔD i0 Indicates the deviation of the inlet outer diameter, ΔE i0 Indicates the inlet ellipticity deviation, ΔT i ΔH represents the change in temperature. i ΔV represents the change in wall thickness. i ΔF represents the velocity fluctuation. ij ΔG represents the change in tension. ij Indicates the roll gap offset, ΔR ij W represents the rolling force deviation. ij The model for predicting dimensional deviations, which represents the amount of wear on the roller system, includes an outer diameter deviation prediction channel and an ellipticity deviation prediction channel. The outer diameter deviation prediction channel is used to calculate and predict the outer diameter deviation based on temperature changes, wall thickness changes, tension changes, roll gap offset, and roller system wear. The ellipticity deviation prediction channel is used to calculate and predict the ellipticity deviation based on temperature changes, wall thickness changes, tension changes, roll gap offset, and roller system wear.

[0028] S3. Input the predicted outer diameter deviation, predicted ellipticity deviation, and deviation source weights into the self-correcting controller. Before the corresponding pipe section reaches the adjustable sizing stand, generate the roll gap compensation and tension compensation to compensate for the predicted outer diameter deviation and predicted ellipticity deviation, and the roll gap offset ΔG. ij Determined according to the following relationship: ΔG ij =G ij f -G ij s -C ij T , Among them, G ij f G represents the roll gap feedback value corresponding to the j-th sizing stand for the i-th pipe segment. ij sC represents the roll gap setting value for the j-th sizing stand. ij T This indicates the amount of heat compensation for the roll gap caused by the temperature rise of the machine frame.

[0029] S4. Based on the running speed of the steel pipe, the current position of the pipe section, and the position of the sizing stand, the roll gap compensation and tension compensation are matched in time and applied when the corresponding pipe section reaches the target sizing stand. In S4, the timing of the timing matching determines the output time of the control command based on the current position of the corresponding pipe section, the position of the target sizing stand, the running speed of the steel pipe, and the response time of the actuator. The output time of the control command is earlier than the time when the corresponding pipe section reaches the target sizing stand.

[0030] S5. Obtain the measured outer diameter and measured ellipticity of the corresponding pipe section after leaving the sizing unit. Compare the measured outer diameter and measured ellipticity with the predicted outer diameter deviation and predicted ellipticity deviation, respectively, to obtain the predicted residual. Update the deviation source weight and the control gain parameter of the self-correcting controller based on the predicted residual. This is used for subsequent pipe section size deviation prediction and self-correcting control. The deviation source weight includes the weight magnitude and the direction of action. The weight magnitude is used to indicate the degree of contribution of the corresponding deviation source to the predicted outer diameter deviation or predicted ellipticity deviation, and the direction of action is used to indicate whether the corresponding deviation source increases or decreases the outer diameter deviation or ellipticity deviation.

[0031] The self-correcting controller includes a deviation judgment unit, a compensation amount generation unit, and a control amount allocation unit. The deviation judgment unit is used to determine whether the predicted outer diameter deviation and the predicted ellipticity deviation exceed the preset deviation threshold. The compensation amount generation unit is used to generate the roll gap compensation amount and the tension compensation amount. The control amount allocation unit is used to determine the allocation ratio of the roll gap compensation amount and the tension compensation amount. When the deviation source weight corresponding to the roll gap offset or roll system wear is greater than the preset equipment disturbance weight threshold, the control amount allocation unit increases the allocation ratio of the roll gap compensation amount; when the deviation source weight corresponding to the tension change is greater than the preset tension disturbance weight threshold, the control amount allocation unit increases the allocation ratio of the roll gap compensation amount.

[0032] The control quantity allocation unit increases the allocation ratio of tension compensation quantity. The roll gap compensation quantity is constrained by the maximum adjustment quantity of roll gap in a single operation, the maximum adjustment difference of roll gap between adjacent pipe sections, and the response time of the roll gap actuator. The tension compensation quantity is constrained by the maximum adjustment quantity of tension in a single operation, the maximum adjustment difference of tension between adjacent pipe sections, and the response time of the tension adjustment mechanism. When the difference in roll gap compensation quantity or tension compensation quantity between adjacent pipe sections exceeds the corresponding maximum adjustment difference, the roll gap compensation quantity or tension compensation quantity is smoothed and limited, so that the control quantity between continuous pipe sections increases or decreases according to the preset change rate.

[0033] The prediction residuals include outer diameter prediction residuals and ellipticity prediction residuals. The outer diameter prediction residuals are determined based on the difference between the measured outer diameter deviation and the predicted outer diameter deviation, while the ellipticity prediction residuals are determined based on the difference between the measured ellipticity deviation and the predicted ellipticity deviation. When the prediction residuals of multiple consecutive pipe segments show a consistent trend with the same source of deviation, the weight of that source of deviation is increased. Conversely, when the prediction residuals of multiple consecutive pipe segments show a inconsistent trend with the same source of deviation, the weight of that source of deviation is decreased.

[0034] Example 2

[0035] Please see Figures 1-3 Based on Example 1, this example further explains the self-correcting control process in step S3. In this example, the self-correcting controller is used to generate roll gap compensation and tension compensation in advance based on the predicted outer diameter deviation, predicted ellipticity deviation, and deviation source weights before the corresponding pipe segment reaches the adjustable sizing stand. This ensures that the actual control action of the sizing stand can be applied to the target pipe segment, rather than performing a delayed correction after the pipe segment has passed.

[0036] Specifically, after the dimensional deviation prediction model outputs the predicted outer diameter deviation and predicted ellipticity deviation of the i-th pipe segment, the self-correcting controller first reads the deviation source weights corresponding to that pipe segment. These deviation source weights include weights for temperature changes, wall thickness changes, tension changes, roll gap offset, and roll wear. The self-correcting controller determines the main cause of the current dimensional deviation based on these weights and determines the allocation ratio of roll gap compensation and tension compensation based on the determination results.

[0037] In this embodiment, the roll gap offset is used to characterize the degree of deviation between the actual roll gap state of the sizing stand and the set roll gap state. Roll gap offset ΔG ij Determined according to the following relationship: ΔG ij =G ij f -G ij s -C ij T , Among them, G ij f G represents the roll gap feedback value corresponding to the j-th sizing stand for the i-th pipe segment. ij s C represents the roll gap setting value for the j-th sizing stand. ij T This represents the roll gap thermal compensation amount caused by the temperature rise of the frame. By introducing the roll gap thermal compensation amount caused by the temperature rise of the frame, we can avoid judging the roll gap offset solely based on the difference between the roll gap feedback value and the roll gap set value, thereby reducing the interference of frame thermal deformation on the compensation amount calculation.

[0038] When ΔG ij When ΔG is positive and its absolute value exceeds the preset roll gap offset threshold, it indicates that the actual roll gap of the j-th sizing stand tends to be larger than the set roll gap. The self-correcting controller uses this roll gap offset as an important basis for generating the roll gap compensation amount. ij When ΔG is negative and its absolute value exceeds the preset roll gap offset threshold, it indicates that the actual roll gap of the j-th sizing stand tends to be smaller than the set roll gap. The self-correcting controller corrects the roll gap compensation amount in the opposite direction based on the predicted outer diameter deviation and the predicted ellipticity deviation. ij When the preset roll gap offset threshold is not exceeded, the self-correcting controller reduces the impact of roll gap offset on the generation of compensation amount, and prioritizes the determination of compensation method based on tension change weight, temperature change weight, or roll system wear weight.

[0039] When generating the compensation amount, the self-correcting controller first determines whether the predicted outer diameter deviation and the predicted ellipticity deviation exceed the corresponding preset deviation thresholds. If neither the predicted outer diameter deviation nor the predicted ellipticity deviation exceeds the preset deviation threshold, the current roll gap and current tension remain unchanged. If either the predicted outer diameter deviation or the predicted ellipticity deviation exceeds the corresponding preset deviation threshold, the compensation amount generation process begins.

[0040] When the roll gap offset weight or roll system wear weight exceeds the preset equipment disturbance weight threshold, the self-correcting controller increases the proportion of roll gap compensation and allocates the roll gap compensation to one or more adjustable sizing stands that the i-th pipe segment is about to reach. The roll gap compensation is used to change the effective roll gap of the target sizing stand, causing the predicted outer diameter deviation and predicted ellipticity deviation to change in a decreasing direction.

[0041] When the tension change weight exceeds the preset tension disturbance weight threshold, the self-correcting controller increases the proportion of tension compensation and applies the tension compensation to the tension control range corresponding to the target sizing stand. The tension compensation is used to adjust the stress state of the pipe section during the sizing process, thus suppressing the outer diameter deviation caused by tension fluctuations in advance.

[0042] When both predicted outer diameter deviation and predicted ellipticity deviation exist simultaneously, the self-correcting controller prioritizes determining whether there is a consistent correction direction based on the deviation directions of the two. If the correction directions corresponding to the predicted outer diameter deviation and predicted ellipticity deviation are consistent, the self-correcting controller generates uniform roll gap compensation and tension compensation amounts according to the weight of the deviation source. If the correction directions corresponding to the predicted outer diameter deviation and predicted ellipticity deviation are inconsistent, the self-correcting controller reduces the single roll gap compensation amount and increases the participation ratio of tension compensation and smoothing limit, avoiding the amplification of one of the dimensional deviations due to a single roll gap adjustment.

[0043] In this embodiment, the roll gap compensation and tension compensation are subject to execution constraints before output. The roll gap compensation does not exceed the maximum roll gap adjustment in a single operation, and the difference in roll gap compensation between adjacent pipe sections does not exceed the maximum adjustment difference in roll gap between adjacent pipe sections. The tension compensation does not exceed the maximum tension adjustment in a single operation, and the difference in tension compensation between adjacent pipe sections does not exceed the maximum adjustment difference in tension between adjacent pipe sections. When the calculated compensation exceeds the above limits, the self-correcting controller limits the compensation to ensure that the compensation is within the allowable range of the actuator.

[0044] In this embodiment, the predicted outer diameter deviation, predicted ellipticity deviation, and deviation source weights are converted into executable roll gap compensation and tension compensation amounts. Since the calculation of the roll gap offset incorporates roll gap feedback values, roll gap setpoints, and frame thermal compensation, the self-correcting controller can distinguish between actual roll gap deviations and measurement deviations caused by frame thermal deformation, thereby improving the targeting of the compensation amount generation. This process transforms dimensional control from passive correction after post-detection to proactive correction before the corresponding pipe section reaches the adjustable sizing frame.

[0045] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes, comprising the following steps, characterized in that: S1. Obtain the incoming material status parameters, sizing machine frame execution parameters, and online detection parameters during the online sizing process of the steel pipe. Based on the steel pipe running speed, sampling time, and the position of each sizing machine frame, divide the steel pipe into continuous pipe segments along the length direction and establish the pipe segment status vector for each segment. S2. Input the pipe segment state vector into the size deviation prediction model, calculate the predicted outer diameter deviation and predicted ellipticity deviation of the corresponding pipe segment before reaching the post-diameter measurement position, and obtain the deviation source weights corresponding to temperature change, wall thickness change, tension change, roll gap offset and roll system wear. S3. Input the predicted outer diameter deviation, predicted ellipticity deviation and deviation source weight into the self-correcting controller, and generate the roll gap compensation amount and tension compensation amount to offset the predicted outer diameter deviation and predicted ellipticity deviation before the corresponding pipe section reaches the adjustable sizing frame. S4. Based on the running speed of the steel pipe, the current position of the pipe section, and the position of the sizing machine stand, the roll gap compensation amount and tension compensation amount are matched in time and applied when the corresponding pipe section reaches the target sizing machine stand; S5. Obtain the measured outer diameter and measured ellipticity of the corresponding pipe segment after leaving the sizing unit. Compare the measured outer diameter and measured ellipticity with the predicted outer diameter deviation and predicted ellipticity deviation, respectively, to obtain the predicted residual. Update the deviation source weight and the control gain parameter of the self-correcting controller based on the predicted residual for subsequent pipe segment size deviation prediction and self-correcting control.

2. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The incoming material status parameters include the target outer diameter of the steel pipe, the target ellipticity, the allowable deviation of the outer diameter, the allowable deviation of the ellipticity, the steel pipe wall thickness, the steel pipe grade, the temperature of the steel pipe before entering the sizing unit, the running speed of the steel pipe, and the outer diameter detection value of the steel pipe before entering the sizing unit. The online detection parameters include the measured outer diameter, measured ellipticity, circumferential distribution value of the outer diameter, and the pipe section detection time collected by the post-diameter measuring device set on the outlet side of the sizing unit. The measured outer diameter and measured ellipticity are used to form the feedback dimension data of the corresponding pipe section.

3. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: When the steel pipe described in S1 is divided into continuous pipe segments along its length, the length of a single pipe segment is determined by the sampling period and the running speed of the steel pipe, and a pipe segment number is assigned to each pipe segment so that the same pipe segment maintains the same pipe segment number when entering the sizing unit, passing through each sizing frame, and reaching the post-diameter measuring position.

4. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The pipe segment state vector is represented as follows: X i =[ΔD i0 ,ΔE i0 ,ΔT i ,ΔH i ,ΔV i ,ΔF ij ,ΔG ij ,ΔR ij ,W ij ], Where i represents the pipe section number, j represents the sizing stand number, and ΔD i0 Indicates the deviation of the inlet outer diameter, ΔE i0 Indicates the inlet ellipticity deviation, ΔT i ΔH represents the change in temperature. i ΔV represents the change in wall thickness. i ΔF represents the velocity fluctuation. ij ΔG represents the change in tension. ij Indicates the roll gap offset, ΔR ij W represents the rolling force deviation. ij This indicates the amount of wear on the roller system.

5. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 4, characterized in that: The roll gap offset ΔG ij Determined according to the following relationship: ΔG ij =G ij f -G ij s -C ij T , Among them, G ij f G represents the roll gap feedback value corresponding to the j-th sizing stand for the i-th pipe segment. ij s C represents the roll gap setting value for the j-th sizing stand. ij T This indicates the amount of heat compensation for the roll gap caused by the temperature rise of the machine frame.

6. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The dimensional deviation prediction model includes an outer diameter deviation prediction channel and an ellipticity deviation prediction channel. The outer diameter deviation prediction channel is used to calculate and predict the outer diameter deviation based on temperature change, wall thickness change, tension change, roll gap offset, and roll system wear. The ellipticity deviation prediction channel is used to calculate and predict the ellipticity deviation based on temperature change, wall thickness change, tension change, roll gap offset, and roll system wear.

7. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The deviation source weight includes a weight magnitude and an action direction. The weight magnitude is used to indicate the degree of contribution of the corresponding deviation source to the predicted outer diameter deviation or the predicted ellipticity deviation, and the action direction is used to indicate whether the corresponding deviation source increases or decreases the outer diameter deviation or ellipticity deviation.

8. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The self-correcting controller includes a deviation judgment unit, a compensation amount generation unit, and a control amount allocation unit. The deviation judgment unit is used to determine whether the predicted outer diameter deviation and the predicted ellipticity deviation exceed the preset deviation threshold. The compensation amount generation unit is used to generate the roll gap compensation amount and the tension compensation amount. The control amount allocation unit is used to determine the allocation ratio of the roll gap compensation amount and the tension compensation amount.

9. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The roll gap compensation amount is constrained by the maximum adjustment amount of the roll gap in a single operation, the maximum adjustment difference of the roll gap between adjacent pipe sections, and the response time of the roll gap actuator. The tension compensation amount is constrained by the maximum adjustment amount of the tension in a single operation, the maximum adjustment difference of the tension between adjacent pipe sections, and the response time of the tension adjustment mechanism.

10. The method for predicting and self-correcting dimensional deviations in the online sizing process of steel pipes according to claim 1, characterized in that: The timing matching described in S4 determines the control command output time based on the current position of the corresponding pipe segment, the position of the target sizing machine, the running speed of the steel pipe, and the response time of the actuator. The control command output time is earlier than the time when the corresponding pipe segment arrives at the target sizing machine.

Images