Cold continuous rolling additional tension self-adaptive dynamic setting method under unsteady rolling condition

Abstract

The invention provides a self-adaptive dynamic setting method for cold continuous rolling additional tension under an unsteady rolling condition, and relates to the technical field of mechanical automation control. The method comprises the following steps: firstly, based on obtained strip steel hot rolling-cold rolling process information, constructing a cross-process cold continuous rolling unsteady state rolling force prediction model; and then an additional tension setting function in the cold continuous rolling unsteady-state speed increasing and decreasing rolling process is constructed with the minimum rolling force fluctuation as the target, finally, the optimal additional tension setting value of each rack is obtained through an artificial intelligence algorithm and is fitted into an additional tension curve, and collaborative setting of the tension of each rack is achieved. Compared with a traditional additional tension setting method, additional tension setting is dynamically adjusted according to the set and actually measured rolling force deviation, speed disturbance in the unstable rolling process can be responded in real time, the rolling stability of multiple low-speed and variable-speed stages is improved, the strip steel head thickness deviation and strip shape defects are effectively reduced, and the yield of the strip steel is improved. And the head quality consistency and the yield of the strip steel are guaranteed.

Description

Technical Field

[0001] This invention relates to the field of mechanical automation control technology, and in particular to an adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions. Background Technology

[0002] The speed-up and down-speed process in cold continuous rolling is a typical unsteady, time-varying process. During this stage, changes in rolling speed synchronously cause changes in the friction coefficient between the rolls and the strip, as well as the strip's deformation resistance, leading to significant fluctuations in rolling force. These fluctuations are a major cause of reduced strip head thickness accuracy and shape defects. They not only disrupt the dynamic stability of the roll gap but also render pre-set compensation methods such as roll bending force ineffective, severely restricting the consistency of product quality across the entire line. To reduce rolling force fluctuations caused by changes in rolling speed, the magnitude of the additional tension between the stands is typically adjusted to offset these changes. This provides crucial stability support for the system during the unsteady rolling stage, thereby improving the thickness and shape control accuracy of the strip during speed changes.

[0003] In traditional cold continuous rolling control, tension compensation is typically adjusted using a fixed tension coefficient during acceleration and deceleration phases. However, relying solely on a single static compensation value is insufficient to adapt to the dynamic characteristics of unsteady rolling processes. It fails to adequately meet the stability requirements of multi-stage low-speed rolling and neglects the tension coupling effect between multiple stands, leading to mutual interference in compensation effects and an inability to effectively suppress rolling force fluctuations. This has long constrained the control accuracy of strip flatness and thickness at the head of the strip. Therefore, a dynamic adaptive tension compensation strategy needs to be developed to respond in real-time to changes in rolling conditions. By optimizing the tension setting in real-time under unsteady conditions, rolling force fluctuations can be minimized, improving overall rolling stability and effectively enhancing strip head thickness and shape accuracy, thus ensuring consistent product quality across the entire line. Summary of the Invention

[0004] To address the aforementioned technical problems in existing technologies, this invention provides an adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions. This method adaptively and dynamically compensates for rolling force fluctuations during the acceleration and deceleration of the strip head in cold continuous rolling. It balances the stability requirements of multi-segment low-speed rolling with the tension coupling effect between multiple stands, minimizing rolling force fluctuations to ensure overall rolling process stability and the accuracy of strip head shape and thickness control. The technical solution is as follows:

[0005] An adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions, the method comprising:

[0006] Step 1: Obtain the hot rolling temperature parameters and the actual rolling force and thickness of the strip steel during cold rolling;

[0007] Step 2: Based on the parameter data obtained in Step 1, construct a high-precision prediction model for unsteady rolling force across processes in cold continuous rolling to predict the cold rolling force;

[0008] Step 3: Establish an additional tension setting function for the unsteady-state speed-up and down-speed rolling process of cold-rolled strip steel with the goal of minimizing rolling force fluctuations;

[0009] Step 4: Determine the constraints of the function in Step 3 based on material properties, smooth rolling, and equipment capacity;

[0010] Step 5: Use artificial intelligence algorithms to obtain the optimal set value of additional tension for each frame, and fit it into an additional tension curve. Based on the curve, realize adaptive dynamic adjustment of additional tension.

[0011] The process of constructing the high-precision prediction model for unsteady rolling force across processes in step 2 of cold continuous rolling is as follows:

[0012] Step 2.1: Based on the strip composition, hot rolling coiling temperature, final rolling temperature and thickness, cold rolling measured tension, thickness and width, and coiling tension, the Prophet model is used to obtain the rolling force and adjustment upper and lower limits of each cold rolling stand, which are used as input features for the LightGBM (Lightweight Gradient Lifter).

[0013] The specific formula in step 2.1 is as follows:

[0014]

[0015] In the formula: The rolling force prediction range; It is the trend term that describes the long-term overall direction of change in a time series; It is a seasonal term that describes periodic changes (week, month, year); These are event items that describe the impact of accidental events; It represents the error term indicating random fluctuations.

[0016] Step 2.2: Optimize the hyperparameters of Lightweight Gradient Boosting Machine (LightGBM) based on Bayesian optimization (BO) algorithm to improve algorithm performance;

[0017] Step 2.3: Use the strip chemical composition, hot rolling finishing temperature and thickness, coiling temperature, cold rolling measured rolling force, thickness, tension and width, and the predicted rolling force value and upper and lower limits of the predicted rolling force value selected from the Prophet model as input features of the optimized LightGBM to predict the cold rolling force.

[0018] The additional tension setting function in step 3 is:

[0019]

[0020] In the formula: For tension compensation; m and j represent the number of strip segments and the number of stands, respectively; g i The weighting coefficient for the rolling force deviation of the i-th segment of the strip is set empirically. This is the predicted value of the strip rolling force in the i-th segment of the j-th stand; The optimal rolling force is the rolling force for the i-th strip on the j-th stand.

[0021] The constraints in step 4 involve material properties, strip slippage, strip narrowing and breakage, and equipment capacity. The constraints are as follows: ; ; ;

[0022] In the formula, The front tension of the i-th segment of the j-th frame; The maximum allowable tension of the i-th segment of the j-th frame is determined by the material yield strength and plate stress. Let be the unit front tension of the j-th frame; , These are the minimum and maximum allowable tension settings under equipment capacity constraints, respectively; μ is the coefficient of friction; α is the bite angle; and p is the rolling force. μ, α, and p are determined by equipment characteristics and material properties.

[0023] The process for obtaining the optimal setpoint for the additional tension of each frame in step 5 is as follows:

[0024] Step 5.1: Divide the strip speed from low speed section to the end of the rolling process into n segments, from low to high speed, and calculate the average speed v at each segment. i,j ;

[0025] Step 5.2: To maintain stable rolling, the tension before the first stand entrance remains essentially constant, and the tension between the fifth stand and the coiler is kept constant. The actual additional tension only exists at exits S1, S2, S3, and S4. Based on the prediction model in Step 2 and the additional tension setting function in Step 3, the hot rolling coiling temperature, hot rolling finishing temperature, cold rolling speed, and unit tension difference variables corresponding to each speed segment point are optimized. The input variable x formulas for the prediction model in Step 2 and the additional tension setting function in Step 3 are as follows:

[0026]

[0027] In the formula, Let be the thickness of the j-th frame under the i-th segment, where i = 1, 2, ..., n-1;

[0028] Let be the average speed of the j-th frame in the i-th segment, where i = 1, 2, ..., n-1;

[0029] Let be the tension difference of the j-th frame under the i-th segment, i=1,2,…,n-1;

[0030] , These represent the thickness, speed, and tension difference at the exit of the j-th frame, respectively.

[0031] T coll,i T col,i These are the hot-rolling coiling temperature and the hot-rolling finishing temperature of the i-th strip, respectively.

[0032] Step 5.3: Using the improved differential evolution-particle swarm optimization algorithm, the optimal solution of the additional tension setting function in step 3 is obtained according to the constraints set in step 4, which is the optimal setting value of the additional tension for each frame.

[0033] In step 5, the optimal set values ​​of additional tension for each frame are fitted into additional tension curves using the least squares method.

[0034] The improved differential evolution-particle swarm optimization algorithm in step 5.3 is as follows:

[0035] A tension compensation strategy is introduced through the three-population algorithm (TPA). The TPA algorithm consists of three populations: one population uses particle swarm optimization (PSO), and the other two populations use two different multi-population differential evolution (DE) modes for optimization. The specific three combination modes are as follows:

[0036] 1) TPA1: Particle Swarm Optimization (PSO) / Randomized Mode (Rand) / Best Mode;

[0037] 2) TPA2: Particle Swarm Optimization (PSO) / Current-best / Randomized Mode (Rand);

[0038] 3) TPA3: Particle Swarm Optimization (PSO) / Current-best / Best.

[0039] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0040] Compared with traditional additional tension setting strategies, the above-mentioned scheme dynamically adjusts the additional tension compensation amount based on different rolling stages and working conditions, with the goal of minimizing rolling force fluctuations. It can respond in real time to speed disturbances in unsteady rolling processes, improve rolling stability in multiple low-speed and variable-speed stages, take into account the tension coupling effect between multiple stands, effectively reduce strip head thickness deviation and strip shape defects, improve head quality consistency and yield while reducing scrap rate. Attached Figure Description

[0041] 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.

[0042] Figure 1 This is a flowchart of an adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions, provided by an embodiment of the present invention.

[0043] Figure 2 This is a flowchart of the calculation process for the additional tension optimization setting in an embodiment of the present invention;

[0044] Figure 3 These are the additional tension setting curves for racks S1 to S5 in this embodiment of the invention, wherein (a) is the additional tension setting curve for rack S1, (b) is the additional tension setting curve for rack S2, (c) is the additional tension setting curve for rack S3, and (d) is the additional tension setting curve for rack S4.

[0045] Figure 4 This is a comparison of rolling force fluctuations in the embodiments of the present invention, wherein (a) is a comparison of rolling force fluctuations before and after optimization of S1 stand, (b) is a comparison of rolling force fluctuations before and after optimization of S2 stand, (c) is a comparison of rolling force fluctuations before and after optimization of S3 stand, (d) is a comparison of rolling force fluctuations before and after optimization of S4 stand, and (e) is a comparison of rolling force fluctuations before and after optimization of S5 stand.

[0046] Figure 5 This is a comparison of the strip thickness difference fluctuation effect before and after the method of the present invention is put into use in the embodiments of the present invention;

[0047] Figure 6 These are typical plate shape diagrams compared in the embodiments of the present invention, wherein (a) is the plate shape before the model is put in, and (b) is the plate shape after the model is put in. Detailed Implementation

[0048] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0049] 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.

[0050] 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.

[0051] 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.

[0052] This invention provides an adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions. For example... Figure 1 The flowchart shown is a method for adaptive dynamic setting of additional tension in cold continuous rolling under unsteady rolling conditions. This method may include the following steps:

[0053] Step 1: Obtain the hot rolling temperature parameters and the actual rolling force and thickness of the strip steel during cold rolling;

[0054] Step 2: Based on the parameter data obtained in Step 1, construct a high-precision prediction model for unsteady rolling force across processes in cold continuous rolling to predict the cold rolling force;

[0055] Step 3: Establish an additional tension setting function for the unsteady-state speed-up and down-speed rolling process of cold-rolled strip steel with the goal of minimizing rolling force fluctuations;

[0056] Step 4: Determine the constraints of the function in Step 3 based on material properties, smooth rolling, and equipment capacity;

[0057] Step 5: Use artificial intelligence algorithms to obtain the optimal set value of additional tension for each frame, and fit it into an additional tension curve. Based on the curve, realize adaptive dynamic adjustment of additional tension.

[0058] The following description, in conjunction with specific embodiments, illustrates this point.

[0059] The additional tension setting method proposed in this invention was implemented using a 1420mm six-roll, five-stand cold rolling mill in a cold rolling plant as an example. The original additional tension setting strategy of this production line was a fixed tension coefficient mode, and a data acquisition platform integrating hot and cold rolling processes was established.

[0060] Perform the calculations as follows.

[0061] Step 1: Obtain the hot-cold rolling process parameters for strip steel, as shown in Table 1; the mill equipment parameters, as shown in Table 2; and the cold continuous rolling production process parameters, as shown in Table 3.

[0062] Table 1 Hot-Cold Rolling Process Parameters for Strip Steel

[0063]

[0064] Table 2 Relevant parameters of cold rolling mill equipment

[0065]

[0066] Table 3 Cold Rolling Production Process Parameters

[0067]

[0068] (Column 5 in the table above is omitted to represent other process parameters, including strip composition, hot-rolled final thickness, rolling speed of stands 1-5, entry thickness, front tension of stands 1-4, back tension of stand 1, coiling tension, etc.)

[0069] Step 2:

[0070] (1) The Prophet model was used to select the predicted rolling force values ​​for each cold rolling stand and to adjust the upper and lower bounds;

[0071] (2) The hyperparameters of Lightweight Gradient Boosting Machine (LightGBM) – number of decision trees and learning rate – were optimized based on the Bayesian optimization algorithm (BO). The optimal values ​​were 213 and 0.19176, respectively.

[0072] (3) such as Figure 2 In the process, LightGBM is used to predict the rolling force and obtain the predicted rolling force values ​​for each stand;

[0073] Step 3: To minimize rolling force fluctuations, the following additional tension setting function is used:

[0074]

[0075] In the formula: m and j represent the number of strip segments and the number of stands, respectively; g i This is the weighting coefficient for the rolling force deviation of the i-th segment of the strip; This is the predicted value of the strip rolling force in the i-th segment of the j-th stand; The optimal rolling force is the rolling force for the i-th strip on the j-th stand.

[0076] Step 4: Based on Table 2 and the smooth rolling process conditions, determine the tension and deviation range limits: ; ; ;

[0077] In the formula, The front tension of the i-th segment of the j-th frame; The maximum allowable tension of the i-th segment of the j-th frame is determined by the material yield strength and plate stress. Let be the unit front tension of the j-th frame; , These are the minimum and maximum allowable tension settings under equipment capacity constraints, respectively; μ is the coefficient of friction; α is the bite angle; and p is the rolling force. μ, α, and p are determined by equipment characteristics and material properties.

[0078] Step 5:

[0079] (1) Divide the speed of the strip from low speed section to the end of the rolling process into n segments, with n being 10. Calculate the average speed at each segment, as shown in Table 4.

[0080] Table 4 Speed ​​Breakpoints for Each Frame

[0081]

[0082] (2) Obtain the strip composition, hot rolling coiling temperature, final rolling temperature and thickness from the data acquisition platform, as well as the measured values ​​of the exit thickness of each cold rolling stand, the measured values ​​of the rolling speed, and the original tension setpoint, according to the following formula:

[0083]

[0084] Determine input variables .

[0085] (3) Using an improved differential evolution-particle swarm optimization algorithm, the additional tension setting function in step 3 is solved according to the constraints set in step 4. The optimal solution is the optimal setting value of the additional tension for each frame. The optimal solution is further fitted into an additional tension curve, such as... Figure 3 .

[0086] The improved differential evolution-particle swarm optimization algorithm is as follows:

[0087] A tension compensation strategy is introduced through the three-population algorithm (TPA). The TPA algorithm consists of three populations: one population uses particle swarm optimization (PSO), and the other two populations use two different multi-population differential evolution (DE) modes for optimization. The specific three combination modes are as follows:

[0088] 1) TPA1: Particle Swarm Optimization (PSO) / Randomized Mode (Rand) / Best Mode;

[0089] 2) TPA2: Particle Swarm Optimization (PSO) / Current-best / Randomized Mode (Rand);

[0090] 3) TPA3: Particle Swarm Optimization (PSO) / Current-best / Best.

[0091] like Figure 4 As shown, after adopting the method proposed in this invention, the rolling force fluctuation rate is reduced to a certain extent, with a fluctuation rate reduction of 3% to 9%, effectively improving the overall stability of the rolling process. Figure 5 and Figure 6 Furthermore, the comparison results of strip head thickness deviation and typical plate shape diagram before and after the optimization of the additional tension setting strategy are given. It can be seen that the additional tension setting strategy given in this invention effectively improves the uniformity of strip head shape, shortens the thickness deviation length, and improves the plate shape quality.

[0092] 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 adaptive dynamic setting of additional tension in cold continuous rolling under unsteady rolling conditions, characterized in that, The method includes: Step 1: Obtain the hot rolling temperature parameters and the actual rolling force and thickness of the strip steel during cold rolling; Step 2: Based on the parameter data obtained in Step 1, construct a high-precision prediction model for unsteady rolling force across processes in cold continuous rolling to predict the cold rolling force; Step 3: Establish an additional tension setting function for the unsteady-state speed-up and down-speed rolling process of cold-rolled strip steel with the goal of minimizing rolling force fluctuations; Step 4: Determine the constraints of the function in Step 3 based on material properties, smooth rolling, and equipment capacity; Step 5: Use artificial intelligence algorithms to obtain the optimal set value of additional tension for each frame, and fit it into an additional tension curve. Based on the curve, realize adaptive dynamic adjustment of additional tension.

2. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 1, characterized in that, The process of constructing the high-precision prediction model for unsteady rolling force across processes in step 2 of cold continuous rolling is as follows: Step 2.1: Based on the strip composition, hot rolling coiling temperature, final rolling temperature and thickness, cold rolling measured tension, thickness and width, and coiling tension, the Prophet model is used to obtain the rolling force and adjustment upper and lower limits of each cold rolling stand, which are used as input features for the LightGBM (Lightweight Gradient Lifter). Step 2.2: Optimize the hyperparameters of Lightweight Gradient Boosting Machine (LightGBM) based on the Bayesian optimization algorithm to improve algorithm performance; Step 2.3: Use the strip chemical composition, hot rolling finishing temperature and thickness, coiling temperature, cold rolling measured rolling force, thickness, tension and width, and the predicted rolling force value and upper and lower limits of the predicted rolling force value selected from the Prophet model as input features of the optimized LightGBM to predict the cold rolling force.

3. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 2, characterized in that, The specific formula in step 2.1 is as follows: ； In the formula: The rolling force prediction range; It is the trend term that describes the long-term overall direction of change in a time series; It is a seasonal term that describes periodic changes; These are event items that describe the impact of accidental events; It represents the error term indicating random fluctuations.

4. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 1, characterized in that, The additional tension setting function in step 3 is: ； In the formula: For tension compensation; m and j represent the number of strip segments and the number of stands, respectively; g i This is the weighting coefficient for the rolling force deviation of the i-th segment of the strip; This is the predicted value of the strip rolling force in the i-th segment of the j-th stand; The optimal rolling force is the rolling force for the i-th strip on the j-th stand.

5. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 1, characterized in that, The constraints in step 4 are as follows: ； ； ； In the formula, The front tension of the i-th segment of the j-th frame; Let be the maximum allowable tension value of the i-th segment of the j-th frame; Let be the unit front tension of the j-th frame; , These represent the minimum and maximum allowable tension settings under equipment capacity constraints, respectively; μ is the coefficient of friction; α is the bite angle; and p is the rolling force.

6. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 1, characterized in that, The process for obtaining the optimal setpoint for the additional tension of each frame in step 5 is as follows: Step 5.1: Divide the strip speed from low speed section to the end of the rolling process into n segments, from low to high speed, and calculate the average speed v at each segment. i,j ; Step 5.2: Based on the prediction model in Step 2 and the additional tension setting function in Step 3, optimize the hot rolling coiling temperature, hot rolling finishing temperature, cold rolling speed, and unit tension difference variables corresponding to each speed segment point. The input variable x formulas for the prediction model in Step 2 and the additional tension setting function in Step 3 are as follows: ； In the formula, Let be the thickness of the j-th frame under the i-th segment, where i = 1, 2, ..., n-1; Let be the average speed of the j-th frame in the i-th segment, where i = 1, 2, ..., n-1; Let be the tension difference of the j-th frame under the i-th segment, i=1,2,…,n-1; , These represent the thickness, speed, and tension difference at the exit of the j-th frame, respectively. T coll,i T col,i These are the hot-rolling coiling temperature and the hot-rolling final rolling temperature of the i-th strip, respectively. Step 5.3: Using the improved differential evolution-particle swarm optimization algorithm, the optimal solution of the additional tension setting function in step 3 is obtained according to the constraints set in step 4, which is the optimal setting value of the additional tension for each frame.

7. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 1, characterized in that, In step 5, the optimal set values ​​of additional tension for each frame are fitted into additional tension curves using the least squares method.

8. The adaptive dynamic setting method for additional tension in cold continuous rolling under unsteady rolling conditions according to claim 6, characterized in that, The improved differential evolution-particle swarm optimization algorithm in step 5.3 is as follows: A tension compensation strategy is introduced through a three-population algorithm, TPA. The TPA algorithm includes three populations, and the three specific combination modes are as follows: 1) TPA1: Particle Swarm Optimization (PSO) / Randomized Mode (Rand) / Best Mode; 2) TPA2: Particle Swarm Optimization (PSO) / Current-best / Rand in Random Mode; 3) TPA3: Particle Swarm Optimization (PSO) / Current-best / Best-best.

Images