A finishing mill leveling automatic reset method based on intelligent memory and zero self-calibration

Abstract

The present application belongs to the technical field of steel metallurgy and industrial automation control, and provides a kind of fine rolling machine leveling automatic reset method based on intelligent memory and zero self-calibration, which records the current fine rolling stand leveling reference value through the one-key trigger instruction of operator;In the subsequent steel coil rolling process, real-time leveling value is collected and offset is calculated;After the stand is thrown, the leveling value is automatically reset to the memory reference based on differential compensation, realizing the cross-steel coil consistency control of leveling initial value;Combined with cylinder zero compensation and roll gap symmetry control, the safety and stability of leveling action are ensured;The present application solves the problems of leveling drift, insufficient steel adaptability, and difficulty in digitizing leveling experience inheritance in the prior art, and can significantly improve the shape consistency, reduce scrap steel and improve the intelligent level of hot continuous rolling line.

Description

Technical Field

[0001] This invention belongs to the field of steel metallurgy and industrial automation control technology, and relates to the leveling control of the finishing mill stand on a hot continuous rolling production line. It provides an automatic reset method for the leveling of the finishing mill based on intelligent memory and zero-point self-calibration. Background Technology

[0002] The finishing mill in a hot strip rolling production line plays a crucial role in controlling the final thickness and shape of the strip. Leveling (single-stand adjustment) of each stand, as a vital component of hydraulic roll gap adjustment, directly impacts the stress balance of the stands, rolling stability, and the quality of the finished strip shape. In current industrial production, initial leveling values ​​are typically set based on operator experience and manually adjusted during rolling via the OS / DS system. Due to significant differences in operator experience, frequent changes in process rhythm, and marked differences in steel properties, leveling parameters are highly susceptible to fluctuation. If the leveling benchmark is not restored or corrected in a timely manner, it will lead to stress imbalance between stands, increased thickness deviation, abnormal strip shape, and even the production of finished strip scrap, significantly impacting production rhythm and product quality.

[0003] While existing Automatic Position Control (APC) systems can compensate for feedback errors in real time, their control logic only applies to the instantaneous leveling process of a single steel coil. They lack the ability to memorize, inherit, and automatically reset leveling parameters across different coils. The leveling reference values ​​cannot be automatically kept consistent across each steel coil, causing the leveling state to continuously deviate with manual operation, making it difficult to establish a long-term stable control strategy. Furthermore, existing control systems generally lack the ability to intelligently analyze, optimize, and select leveling values, and have not established a parameter classification and management mechanism for different steel grades.

[0004] In actual production, existing technologies do not consider the fine-tuning changes of leveling values ​​during the rolling process. Operator leveling actions are often temporary and random, lacking an automated mechanism, preventing the production line from accumulating and inheriting leveling strategies over long cycles and across steel coils. Existing systems also lack an automatic leveling reset mechanism after steel discarding, failing to automatically restore the optimized leveling benchmark to the next steel coil, leading to leveling drift during production.

[0005] In summary, existing leveling control technologies have the following shortcomings: a lack of memory and automatic reset mechanisms for leveling parameters; the inability to inherit leveling parameters across steel coils; a lack of leveling optimization and intelligent optimization capabilities; and the absence of a unified intelligent link between leveling and hydraulic cylinder compensation. These technical deficiencies restrict the development of automation levels in finishing mills and intelligent rolling technology, urgently requiring a more intelligent, adaptive, and learnable leveling control method. Summary of the Invention

[0006] The purpose of this invention is to address the long-standing problems in the leveling control of hot strip mill finishing mills, such as unstable leveling benchmarks, inability to inherit leveling parameters across coils, and frequent and fluctuating manual adjustments. This invention provides an automatic reset method for finishing mill leveling based on intelligent memory and zero-point self-calibration, thereby improving leveling stability, reducing manual reliance, lowering scrap rate, and comprehensively enhancing the intelligence level and product quality consistency of hot strip mill production lines.

[0007] The technical solution adopted by the present invention to achieve the above objectives is as follows: An automatic reset method for leveling a finishing mill based on intelligent memory and zero-point self-calibration includes: S1. Set up manual one-click trigger to collect the leveling offset or cylinder displacement of each stand of the finishing mill in real time, and store the leveling offset or cylinder displacement as the leveling reference value a. S2. Define the differential compensation and automatic reset algorithm based on offset, specifically including: S21. Collect the real-time leveling value b during the rolling of the next piece of steel; S22. Calculate the offset c = ba; S23. Perform reverse compensation -c after a delay of T seconds following the steel throwing on the frame; S3. Combine the zero-position memory value with the calculated roll gap reference value, and use the upper and lower hydraulic cylinders to compensate for the offset to maintain roll gap symmetry. S4. An interlocking mechanism is set between the automatic leveling APC and the manual one-key trigger automatic leveling reset. When the manual one-key trigger automatic leveling reset signal is detected, the automatic leveling APC switches to the manual trigger automatic leveling reset priority mode, and automatically restores the automatic leveling control strategy after the manual one-key trigger automatic leveling reset ends.

[0008] Furthermore, the offset compensation setting includes bidirectional limiting protection for both maximum and minimum values ​​to ensure that the cylinder's movement is within the mechanically permissible range.

[0009] Furthermore, by using position sensors and system status detection devices, the zero-point reference parameters of the roll gap are collected in real time and dynamically updated during equipment operation, realizing online zero-point self-calibration control.

[0010] Furthermore, each rack is independently leveled, and multi-rack synchronous coordinated control is achieved through unified scheduling logic.

[0011] Furthermore, the manual one-button triggering automatic leveling reset method also includes setting button state lock, continuous trigger jitter protection, and an automatic exit mechanism for abnormal states.

[0012] The beneficial effects of this invention are: Compared with existing finishing mill leveling control technology, it has made significant substantial progress and has beneficial effects, mainly reflected in the following aspects: First, this method achieves digital storage of leveling benchmark values ​​through a "one-click leveling parameter memory mechanism," enabling the system to actively record and inherit manual experience across steel coils. This solves the problem of traditional leveling benchmarks constantly drifting with manual operation, effectively avoiding abnormal plate shape and scrap steel caused by accumulated deviations in leveling parameters. This memory mechanism transforms "unstable operational experience" into "retrievable digital parameters," laying a stable benchmark for leveling control.

[0013] Secondly, this method introduces for the first time the "differential compensation reset algorithm based on offset". By calculating the difference between the memory leveling value and the real-time leveling value, the leveling recovery is automatically performed after the steel is thrown from the frame, so that the leveling value accurately returns to the optimal state in the next piece of steel, achieving true "consistency control of initial leveling value".

[0014] Third, this method ensures the accuracy and safety of roller gap adjustment through measures such as hydraulic cylinder zero-position compensation, OS / DS coordinated leveling, and upper and lower limit safety protection, making leveling more stable and controllable.

[0015] In summary, this method achieves intelligent leveling control throughout the entire process of "memory-recovery-optimization-genetics-closed-loop optimization," which is unattainable by traditional leveling methods. It significantly improves rolling stability and strip shape consistency, reduces scrap rate, and reduces reliance on operational experience, bringing significant benefits to the intelligent upgrading of hot strip rolling production lines. Detailed Implementation

[0016] The present invention will be further described below with reference to embodiments: An automatic reset method for leveling a finishing mill based on intelligent memory and zero-point self-calibration includes: S1. Set up a manual one-click trigger leveling memory / reset button to collect the leveling offset or cylinder displacement of each stand in the current finishing mill in real time, including the leveling offset or cylinder displacement of the upper and lower sides, and store the leveling offset or cylinder displacement as the leveling reference value a for subsequent steel coil leveling start reference.

[0017] The introduction of this mechanism allows the operator's experience-based leveling results to be digitized, standardized, and stored across steel coils, avoiding the problem of the initial leveling value constantly shifting after multiple adjustments, thus ensuring the stability of leveling control from the source.

[0018] Meanwhile, each rack establishes independent leveling control, and multi-rack synchronous coordinated control is achieved through unified scheduling logic.

[0019] S2. Define the differential compensation and automatic reset algorithm based on offset, specifically including: S21. During the rolling of the next piece of steel after recording the leveling reference value a, collect the real-time leveling value b. S22. Perform a difference calculation on a and b, and calculate the offset of the leveling value relative to the reference value c=ba; S23. After the steel is thrown from the frame, apply reverse compensation -c to the offset c to restore the leveling value to the recorded state, so as to achieve automatic reset and alignment with the reference. This compensation is automatically executed after each frame throws steel in sequence and after a 2-second delay.

[0020] Unlike existing automatic leveling (APS) models, this method's differential compensation does not rely on real-time plate shape error, but rather on manual experience benchmarks to achieve accurate recovery, ensuring consistency of the leveling benchmark across multiple steel coils and preventing deviations caused by accumulated leveling drift.

[0021] S3. Combine the zero-position memory value, i.e. the installation calibration value, with the calculated roll gap reference value, and use the upper and lower cylinders to compensate for the offset to maintain roll gap symmetry; set a bidirectional limit constraint mechanism for maximum and minimum values ​​to prevent the actuator from overtravel or abnormal amplification, ensuring that the cylinder action is within the mechanical allowable range and improving safety.

[0022] Meanwhile, the zero-point reference parameters of the roll gap are collected in real time through position sensors and system status detection devices, and are dynamically updated during equipment operation to achieve online zero-point self-calibration control.

[0023] S4. An interlocking mechanism is set between the automatic leveling APC and the manual one-key trigger automatic leveling reset. When the manual one-key trigger automatic leveling reset signal is detected, the automatic leveling APC switches to the manual trigger automatic leveling reset priority mode, and automatically restores the automatic leveling control strategy after the manual one-key trigger automatic leveling reset ends.

[0024] The manual one-button triggering automatic reset method also includes button status locking, continuous trigger jitter protection, and an automatic exit mechanism for abnormal states, ensuring the entire leveling process is safe, stable, and reliable.

[0025] This method, under the automatic operation of the finishing mill, triggers a monotonic reset command through a human-machine interface terminal, collects the current manual leveling settings of each stand, and latches and stores them to form a set of leveling reference parameters, thereby realizing intelligent memory of the leveling reference.

[0026] This method uses position sensors to collect the zero-position reference value of the roll in real time, forming a set of zero-point reference parameters, and dynamically updates them according to the equipment's operating status, thus forming a zero-point self-calibration mechanism.

[0027] This method uses floating-point addition to fuse the leveling reference parameters and the zero-point reference parameters to generate a dynamic leveling target value as a real-time control reference, thus forming a reference fusion calculation mechanism.

[0028] This method collects the actual roll gap position feedback value of each frame in real time, performs floating-point difference calculation with the dynamic leveling target value, generates the leveling deviation, forms the basis of closed-loop control, and forms a deviation closed-loop control mechanism.

[0029] This method performs bidirectional amplitude limiting on the leveling deviation, restricting the compensation amount to within the preset safe upper and lower limits, forming a safe amplitude limiting protection mechanism to effectively prevent overtravel and overcompensation problems.

[0030] After detecting the signal that the strip has been thrown out, this method triggers a preset time delay and outputs a compensation command to the hydraulic roll gap actuator in the opposite direction to the leveling deviation, so that the frame leveling value automatically returns to the memory reference state, forming a reverse compensation automatic reset mechanism.

[0031] This method also establishes a historical database of leveling parameters. Based on steel type information, plate thickness specifications, and quality evaluation results, the optimal leveling benchmark value is selected through rule optimization algorithm or weight adaptive algorithm, thereby realizing the digital inheritance and continuous optimization of operational experience.

[0032] Taking the leveling of the stands of the company's hot continuous rolling mill F1-F7 as an example S1. Set a one-key start button on the control panel. When the button is pressed, the finishing mill starts the monotonic holding function and temporarily stores the monotonic values ​​of the current F1-F7 stand rolling mill. When the button is pressed again, the monotonic memory values ​​are cleared and the original control timing is restored.

[0033] S2. When the finishing strip is finished, that is, after each stand throws the steel, the monotonic reset function is gradually started after a 2-second delay. The monotonic value of each finishing stand is automatically reset to the saved monotonic value when the button is pressed.

[0034] The above functions are effective in automatic finishing rolling mode, but not in non-automatic mode; the finishing rolling monotonic reset function is effective throughout the rolling process.

Claims

1. A method for automatic reset of a finishing mill leveling based on intelligent memory and zero self-calibration, characterized in that: include: S1. Set up manual one-click trigger to collect the leveling offset or cylinder displacement of each stand of the finishing mill in real time, and store the leveling offset or cylinder displacement as the leveling reference value a. S2. Set up a differential compensation-based automatic reset algorithm based on offset, specifically including: S21. Collect the real-time leveling value b during the rolling of the next piece of steel; S22. Calculate the offset c = ba; S23. Perform reverse compensation -c after a delay of T seconds following the steel throwing on the frame; S3. Combine the zero-position memory value, i.e. the installation calibration value, with the calculated roll gap reference value, and use the upper and lower hydraulic cylinders to compensate for the offset and maintain the symmetry of the roll gap. S4. An interlocking mechanism is set between the automatic leveling APC and the manual one-key trigger automatic leveling reset. When the manual one-key trigger automatic leveling reset signal is detected, the automatic leveling APC switches to the manual trigger automatic leveling reset priority mode, and automatically restores the automatic leveling control strategy after the manual one-key trigger automatic leveling reset ends.

2. The automatic reset method for leveling a finishing mill based on intelligent memory and zero-point self-calibration according to claim 1, characterized in that: The offset compensation setting described in S3 provides bidirectional limiting constraint protection for both maximum and minimum values, ensuring that the cylinder's movement is within the mechanically permissible range.

3. The automatic reset method for leveling a finishing mill based on intelligent memory and zero-point self-calibration according to claim 1, characterized in that: By using position sensors and system status detection devices, the zero-position reference parameters of the roll gap are collected in real time and dynamically updated during equipment operation, realizing online zero-position self-calibration control.

4. The automatic reset method for leveling a finishing mill based on intelligent memory and zero-point self-calibration according to claim 1, characterized in that: Each rack has its own independent leveling control, and multi-rack synchronous coordinated control is achieved through unified scheduling logic.

5. The automatic reset method for leveling a finishing mill based on intelligent memory and zero-point self-calibration according to claim 1, characterized in that: The manual one-button triggering automatic leveling reset method also includes setting button status lock, continuous trigger jitter protection, and abnormal state automatic exit mechanism.