Cross-process flattening mill rolling force calculation method, device, system and storage medium

Abstract

The application discloses a cross-process skin pass mill rolling force calculation method and device, system and storage medium, and comprises the following steps: S1, a cold continuous rolling process control model is constructed; S2, a deformation resistance data set is obtained according to the cold continuous rolling process control model; and S3, skin pass rolling force calculation is performed through cross-process data fusion based on the deformation resistance data set. The technical scheme of the application solves the problem of low skin pass rolling force precision in the prior art, thereby significantly improving the shape quality of the plate strip steel product.

Description

Technical Field

[0001] This invention belongs to the field of plate and strip steel rolling production technology, and particularly relates to a method, device, system, and storage medium for calculating the rolling force of a leveling machine across multiple processes. Background Technology

[0002] Leveling, a crucial process in strip steel production, directly impacts the surface quality, mechanical properties, and shape of the finished product. After cold rolling and annealing, strip steel typically undergoes a small deformation reduction during leveling rolling to eliminate the yield plateau (generally, the elongation after leveling is less than 3%), thereby reducing or eliminating the Lüders band phenomenon during stamping. Therefore, the accuracy of the leveling rolling force setting directly affects the shape of the load-bearing roll gap, which in turn affects the strip threading stability of the mill and the shape, mechanical properties, and surface roughness of the finished strip. Because the reduction in leveling rolling is very small, its deformation mechanism differs significantly from that of ordinary rolling, making it difficult for conventional rolling force models to accurately predict the leveling rolling force. Currently, modeling methods for leveling rolling force can be categorized into three types: mechanistic methods, intelligent methods, and methods combining mechanistic or intelligent models with parameter self-learning. Strip and sheet production is a typical process industry. Traditional hot rolling and cold rolling processes and production lines are relatively independent systems. Although process and quality data are collected for each production process, multi-dimensional process quality information within each process is still scattered in different systems, and the information cannot be effectively utilized. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method, device, system and storage medium for calculating the rolling force of a leveling machine across processes, thereby solving the problem of low accuracy of leveling rolling force in the prior art and significantly improving the shape quality of strip steel products.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A method for calculating the rolling force of a leveling mill that spans multiple processes includes:

[0006] Step S1: Construct a cold continuous rolling process control model; wherein, the cold continuous rolling process control model includes: rolling force model, friction coefficient model and deformation resistance model;

[0007] Step S2: Obtain the deformation resistance dataset based on the cold continuous rolling process control model;

[0008] Step S3: Based on the deformation resistance dataset, obtain the leveling rolling force calculation model based on cross-process data fusion. At the same time, input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force.

[0009] As a preferred option, in step S2, based on the actual production data of cold rolling mill, the rolling force model, friction coefficient model and deformation resistance model are deformed by inverse calculation based on the deformation resistance of cold rolling mill to obtain the average actual deformation resistance of each stand of cold rolling mill.

[0010] Preferably, step S3 includes:

[0011] The actual deformation resistance of the strip is obtained by inverse solving the average actual deformation resistance of each stand in the cold continuous rolling mill.

[0012] Based on the average actual deformation resistance of each stand in cold rolling mill and the actual deformation resistance of strip steel, a calculation model for leveling rolling force based on cross-process mechanism-data fusion is constructed.

[0013] The actual production data of cold continuous rolling is input into the leveling rolling force calculation model to predict the rolling force.

[0014] The present invention also provides a cross-process leveling mill rolling force calculation device, comprising:

[0015] The building module is used to construct the cold continuous rolling process control model; the cold continuous rolling process control model includes: rolling force model, friction coefficient model and deformation resistance model;

[0016] The calculation module is used to obtain the deformation resistance dataset based on the cold continuous rolling process control model;

[0017] The prediction module is used to obtain a leveling rolling force calculation model based on cross-process data fusion from the deformation resistance dataset, and at the same time input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force.

[0018] As a preferred option, the calculation module uses the actual production data of cold rolling mill to deform the rolling force model, friction coefficient model and deformation resistance model through inverse operation based on the deformation resistance of cold rolling mill, so as to obtain the average actual deformation resistance of each stand of cold rolling mill.

[0019] Preferably, the prediction module includes:

[0020] The calculation unit is used to obtain the actual deformation resistance of the strip by inverse solving the average actual deformation resistance of each stand in the cold rolling mill;

[0021] The building unit is used to construct a leveling rolling force calculation model based on cross-process mechanism-data fusion, according to the average actual deformation resistance of each stand in cold continuous rolling and the actual deformation resistance of the strip.

[0022] The prediction unit is used to input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force.

[0023] This invention also provides a cross-process leveling mill rolling force calculation system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program, when executed by the processor, performs a cross-process leveling mill rolling force calculation method.

[0024] This invention also provides a storage medium storing a computer program that, when running, executes a cross-process leveling mill rolling force calculation method.

[0025] This invention establishes a deformation resistance back-calculation model by numerical simulation modeling and accumulating production data, using actual data from the upstream cold rolling process. Based on this, the deformation resistance of cold rolling is applied to the leveling mill to establish a cross-process leveling rolling force calculation model, which solves the problem of low leveling rolling force accuracy in the existing technology, thereby significantly improving the plate shape quality of strip steel products. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0027] Figure 1 This is a flowchart of the method for calculating the rolling force of a leveling mill across multiple processes according to an embodiment of the present invention;

[0028] Figure 2 This is a schematic diagram showing the division of the work roll and the strip mesh, where, Figure 2 (a) is a schematic diagram of the overall grid division of the work roll. Figure 2 (b) is a schematic diagram of the local grid division of the work roll. Figure 2 (c) is a schematic diagram of the grid division at the contact position between the work roll and the strip;

[0029] Figure 3 A schematic diagram illustrating the vertical support reaction force at the center of the work roll during stabilization rolling;

[0030] Figure 4 This is a schematic diagram of the flattened shape after flattening;

[0031] Figure 5 Schematic diagram of contact arc length under different strip elongation;

[0032] Figure 6 Schematic diagram of contact arc length under different strip yield strengths;

[0033] Figure 7Schematic diagram of contact arc length under different friction coefficients;

[0034] Figure 8 Schematic diagram of contact arc length for different working roll diameters;

[0035] Figure 9 Schematic diagram of contact arc length for different thicknesses;

[0036] Figure 10 Schematic diagram of contact arc length under different tensions;

[0037] Figure 11 This is a schematic diagram of the particle swarm optimization algorithm.

[0038] Figure 12 This is a schematic diagram comparing the actual rolling force with the predicted rolling force. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] Example 1:

[0042] like Figure 1 As shown, an embodiment of the present invention provides a method for calculating the rolling force of a leveling mill across multiple processes, including:

[0043] Step S1: Establish a cold continuous rolling process control model

[0044] 1) Rolling force model

[0045] Rolling force is a core component of cold continuous rolling process settings. The calculation results of the rolling force model are the basis for roll gap settings; therefore, the accuracy of the rolling force model directly affects the thickness and shape quality of the strip. Furthermore, rolling force settings are also fundamental to some settings calculations in cold continuous rolling process control, impacting the entire rolling cycle. This paper introduces the Bland-Ford-Hill model as the rolling force model, whose form is shown in the equation.

[0046]

[0047] Where P is the rolling force and B is the strip width. For the dynamic deformation resistance of the i-th frame, k T Q is the tension influence coefficient. s R' is the stress state coefficient; i H is the flattening radius of the work roll of the i-th frame. i h is the thickness at the entrance of the i-th rack. i Let be the thickness of the outlet of the i-th rack.

[0048] The stress state coefficients of the model are expressed as follows:

[0049]

[0050] in, μ is the reduction rate of the i-th rack pass; i is the coefficient of friction.

[0051] The formula for calculating the reduction rate per pass is shown in the figure.

[0052]

[0053] The flattening radius of the work roll can be calculated using the following formula, which contains both rolling force and flattening radius. It can be achieved through iterative or explicit calculation.

[0054]

[0055] Among them, v i Let E be the rolling speed of the i-th stand, and E be the elastic modulus of the work roll.

[0056] Tension is a non-negligible factor in the cold rolling process. The tension factor in the rolling force model is expressed by the tension influence coefficient, as shown in the equation.

[0057]

[0058] in, The front tension of the i-th frame, Let be the back tension of the i-th frame; Let be the dynamic deformation resistance of the i-th frame.

[0059] 2) Friction coefficient model

[0060] In cold continuous rolling, the friction state between the strip and the work rolls is expressed by the friction coefficient. Factors influencing the friction coefficient include the surface conditions of the strip and rolls, the rolling lubrication state, and the rolling speed. In the relevant setting calculations for the cold continuous rolling process control level, the accuracy of the friction coefficient calculation affects the accuracy of the rolling force calculation; therefore, the setting of the friction coefficient has a significant impact on product quality. The friction coefficient model used in this invention is shown in the following equation.

[0061]

[0062] Where μ0~μ6 are the model parameters of the modal friction coefficient; N r This refers to the number of rolls produced after the roll change.

[0063] 3) Deformation resistance model

[0064] Deformation resistance is a crucial control parameter in the cold rolling process. As a key module in rolling force model calculations, it is also a core component of high-precision strip shape control calculations. While the deformation resistance of strip steel primarily depends on the material's yield strength, cold rolling is a three-dimensional plastic deformation process involving multiple processing steps. Therefore, the deformation resistance in cold rolling is influenced by factors such as the degree and speed of deformation, and cannot be directly characterized by the yield strength. Furthermore, it is also affected by the work hardening effect. Thus, the deformation resistance in cold rolling is a complex parameter that integrates material properties and process characteristics.

[0065] One of the main factors affecting deformation resistance is the degree of material deformation. Based on the production mode of cold continuous rolling process, several reduction rates of the material are first defined, as shown in the formula.

[0066]

[0067] in, The reduction rate before rolling on the i-th stand; Let i be the reduction rate after rolling on the i-th stand; Let a be the average reduction rate of the i-th frame; r This is an empirical value, usually taken as 0.75.

[0068] Based on the above reduction rates, the static deformation resistance is as shown in the formula.

[0069]

[0070] in, Let be the static deformation resistance of the i-th frame, k0 be the baseline value of the deformation resistance, which is related to the inherent properties of the material; m and n are model coefficients, whose values ​​are related to the material properties and the cold rolling work hardening effect.

[0071] To make the deformation resistance model more consistent with actual rolling conditions, the influence of deformation rate is added in addition to considering the degree of deformation. The deformation resistance calculated considering the deformation rate is called dynamic deformation resistance, which is now widely used in cold continuous rolling process settings. The deformation rate can be calculated according to the S. Ekelend formula, as shown in the figure.

[0072]

[0073] Where, ε i Let be the deformation rate of the i-th frame.

[0074] Combining the above equations, we can obtain the formula for calculating dynamic deformation resistance, as shown in the equation.

[0075]

[0076] Where η is the strain rate sensitivity coefficient.

[0077] Step S2: Establish a deformation resistance dataset based on the inverse operation of cold continuous rolling deformation resistance.

[0078] 1) Collect and process cold rolling data

[0079] The cold rolling mill data platform is used to obtain actual historical production data of the cold rolling mill. The data mainly includes hot-rolled incoming coil number, cold-rolled exit coil number, steel grade, entrance thickness, entrance width, and the reduction rate, rolling force, and entrance tension of each stand. This data is preprocessed by deleting null values; multiple rows of identical data exist, but since each strip has a different exit coil number, redundant data can be deleted by filtering by exit coil number; data that is obviously illogical is processed, and the processed data is saved to prepare for building a deformation resistance prediction model.

[0080] 2) Inverse calculation of cold continuous rolling deformation resistance

[0081] In the current cold continuous rolling process, there is no effective means to detect the actual deformation resistance of strip steel during the rolling process on each stand. However, actual deformation resistance is an indispensable data point in the process of establishing a deformation resistance prediction model. The actual deformation resistance can be obtained through inverse deformation resistance calculation. The principle is to use the measured rolling force as a known quantity in the rolling force model, replace other process data with measured values, and use the deformation resistance as an unknown quantity to perform inverse calculation to obtain the actual deformation resistance.

[0082] Based on the deformation resistance model and rolling force model described in the previous section, the rolling force is taken as a known quantity and the deformation resistance as an unknown quantity, and the rolling force formula is separated to obtain the following formula.

[0083]

[0084] in, For the actual deformation resistance, k' T P represents the measured tension coefficient. mea To measure the rolling force, Q' S The stress state coefficient is the measured thickness. The inlet and outlet thicknesses were measured.

[0085] Expanding the tension coefficient on the left side of the above equation and setting the right side of the equation to A1, we obtain the following equation.

[0086]

[0087] in, For the actual measured tension, The measured tension.

[0088] For the above formula After separation and arrangement, the inverse operation model of deformation resistance is obtained, as shown in the equation.

[0089]

[0090] By back-calculating the production data of the cold rolling mill using this formula, the average actual deformation resistance of each strip in each stand can be obtained, which serves as the data resource basis for establishing the next prediction model.

[0091] Step S3: Calculation of leveling rolling force based on cross-process mechanism-data fusion

[0092] In actual production, the deformation mechanism of leveling rolling differs significantly from that of cold rolling. Traditional rolling force models struggle to accurately predict leveling rolling force, and large deviations between the set and actual rolling force lead to fluctuations in the elongation of the strip, negatively impacting surface roughness and shape control. Domestic and international scholars have developed various leveling rolling force calculation models. For example, Roberts WL developed an explicit calculation model for leveling rolling force based on the characteristics of the leveling process. However, this model's derivation is based on a large reduction rate and cannot be directly applied to rolling conditions with low elongation. The rolling force calculated using this formula has a significant error compared to the actual rolling force. Therefore, this study analyzes the reasons for the calculation error in Roberts' leveling rolling force formula. Furthermore, deformation resistance is a crucial factor affecting leveling rolling force during the leveling rolling process, but it cannot be accurately obtained. This study considers optimizing the calculation model by combining cross-process data with actual production data, establishing a cross-process mechanism-data fusion-based leveling rolling force calculation model.

[0093] 1) Classical model for calculating rolling force during leveling

[0094] Roberts' formula for calculating leveling rolling force is as follows:

[0095] P = fB

[0096]

[0097] Where P is the rolling force, f is the rolling force per unit width, B is the strip width, L is the contact arc length between the roll and the strip in the rolling deformation zone, D is the work roll diameter, ε is the strip elongation, μ is the coefficient of friction, h0 is the inlet thickness, and σ p For deformation resistance, v is the strain rate, a is the rolling speed, and σ is the strain rate coefficient. s This represents the yield strength measured under conditions of very low tensile strain rates.

[0098] We selected actual production data from a certain leveling mill unit. The selected data range is shown in the table below. We selected 49 sets of data from this unit that fell within the above range. Substituting the selected data into Roberts' leveling rolling force calculation formula and comparing it with the actual rolling force in production, the result obtained by Roberts' calculation model differed from the actual leveling rolling force data by approximately 6.9 times.

[0099] Table 1

[0100] parameter Numerical range Elongation / % 0.7 Yield strength / MPa 180 Thickness range / mm 0.6—0.7 Width range / mm 1500—1600 <![CDATA[Rolling speed range / m·min -1 > 101—109 Inlet tension / kN 34—45.6 Outlet tension / kN 40—50.64

[0101] Based on the actual leveling rolling force, the Roberts formula was used for back-calculation, yielding an average contact arc length of 9.2 mm. However, this value differs significantly from the empirical value of the contact arc length between the work roll and the strip during the leveling process. The accuracy of the contact arc length calculation during leveling is directly related to the accuracy of the leveling rolling force calculation and is a key factor in ensuring the accuracy of the rolling force calculation.

[0102] 2) Two-dimensional finite element model of the leveling rolling process

[0103] As mentioned in the previous section, the traditional formula for calculating the contact arc length in the leveling process has a large error. Therefore, the simulation of the leveling process will be used to calculate the contact arc length between the work roll and the strip under different working conditions.

[0104] ① Model simplification and assumptions

[0105] Assuming the strip is an ideal elastoplastic body with a finite length, and that the rolling process is symmetrical in the vertical direction, a 1 / 2 model is established, with the work rolls assumed to be elastic bodies. Table 2 shows the material properties of the work rolls and the strip.

[0106] Table 2

[0107] parameter numerical values Work roll elasticity model / MPa 225000 Poisson's ratio of working rolls 0.3 working roll density <![CDATA[7.85×10 3 ]]> Elastic model of strip steel / MPa 206000 Poisson's ratio of strip steel 0.3 strip density <![CDATA[7.85×10 3 ]]>

[0108] ② Interaction settings

[0109] The work roll and the strip are initially tangent, and their interaction is a surface-to-surface contact, where the work roll surface is the master surface and the strip surface is the slave surface. Normal contact is hard contact, and tangential contact is penalized. To facilitate applying loads at the center of the work roll, a small circle with a radius of 25 mm is drawn at the center of the work roll and constrained as a rigid body.

[0110] ③ Analysis Step Settings

[0111] The simulation process consists of three analysis steps: applying tension, pressing down the work roll, and rolling. The analysis step type is static general, and geometric nonlinearity is enabled.

[0112] ④ Loads and boundary conditions

[0113] A reference point RP is established at the center of the work roll, and a displacement perpendicular to the strip is applied downwards. A symmetrical constraint is applied vertically to the lower surface of the strip, and normal loads are applied to the head and tail end faces of the strip to simulate the tension on the strip. An angle is applied at the reference point RP at the center of the work roll to simulate the rotation of the work roll.

[0114] ⑤ Mesh generation and element selection

[0115] The work roll and strip unit type adopt CPE4R. The work roll grid size is larger than the strip grid size. For example... Figure 2 Figure (a) shows a schematic diagram of the mesh generation for the overall structure of the work roll. In the model, only the 30° sector of the contact area between the work roll and the strip is finely meshed to improve computational efficiency. Figure 2 As shown in (b); the grid size of the contact area between the work roll and the strip is 0.1 × 0.1 mm, and the grid size of the strip is 0.05 × 0.05 mm. The grid division of the contact area between the work roll and the strip is as follows: Figure 2 As shown in (c), the total number of elements in the entire finite element model is 168176.

[0116] 3) Two-dimensional finite element simulation study on the flattening deformation behavior of the work roll

[0117] ① Analysis of the elastic-plastic deformation of the work rolls and strip in the rolling zone

[0118] A set of working conditions in the leveling and rolling process of the leveling mill was selected for simulation. The working roll diameter is 650mm, the working roll linear speed is 80m / min, the strip entry thickness is 0.5mm, the strip elongation is 2.8%, the yield strength is 400MPa, the tension on the strip is 40.2MPa, and the friction coefficient between the working roll and the strip is 0.12.

[0119] During stable rolling of the model, the average rolling force per unit plate width is 3188N, as follows: Figure 3 As shown, the surface displacement of the strip in the rolling zone is as follows: Figure 4 As shown.

[0120] ② Influence of strip yield strength on contact arc length

[0121] The parameters that affect the contact arc length during the simulation process are shown in Table 3.

[0122] Table 3

[0123]

[0124] like Figures 5 to 6 As shown, with the increase of strip elongation during the leveling process, the contact arc length increases almost linearly. When the elongation is 0.8%, the contact arc length is 5.5 mm, and when the elongation is 3%, the contact arc length is 6.47 mm. With the increase of strip yield strength, the contact arc length increases. When the yield strength increases from 250 MPa to 550 MPa, the contact arc length increases from 5.61 mm to 11.25 mm.

[0125] like Figures 7 to 8 As shown, with the increase of the friction coefficient between the work roll and the strip, the contact arc length increases. The friction coefficient increases from 0.06 to 0.15, and the contact arc length increases from 4.23 to 6.3 mm. With the increase of the diameter of the work roll, the contact arc length also increases. The diameter of the work roll increases from 600 mm to 650 mm, and the contact arc length increases from 3.84 to 5.61 mm.

[0126] like Figures 9 to 10 As shown, with the increase of strip thickness, the contact arc length increases. When the thickness increases from 0.3 mm to 1.2 mm, the contact arc length increases from 5.4 mm to 6.25 mm. With the increase of tension, the contact arc length decreases. When the tension increases from 0 to 30% of the yield strength, the contact arc length decreases from 6.59 mm to 5.31 mm.

[0127] ③ Data-mechanism fusion model of leveling rolling force

[0128] Based on the influence of parameters such as strip thickness, work roll diameter, and elongation on the contact arc length in the previous section, the contact arc length formula in Roberts' leveling rolling force formula is reconstructed as shown in the following formula.

[0129] P = fB

[0130]

[0131] Table 4 lists the names of the parameters and the units used in the calculations. All units are in the International System of Units (SI).

[0132] Table 4

[0133]

[0134] Where k0-k8 are the coefficients to be optimized in the contact arc length formula.

[0135] ④ Cross-process mechanism - data fusion-based calculation model for leveling rolling force

[0136] In the original formula for calculating the leveling rolling force, the value of deformation resistance is a relatively fixed value for the same type of strip steel. However, in reality, the deformation resistance of strip steel is affected by a variety of factors, such as deformation temperature, deformation rate, and degree of deformation. Strip steel of the same type of steel will produce different deformation resistances. Therefore, the above formula cannot reflect the actual deformation resistance of strip steel.

[0137] To reflect the true deformation resistance of the strip steel, the cold rolling process preceding the leveling process is selected. Using actual production data from the cold rolling process, the deformation resistance of the strip steel that conforms to the actual deformation resistance is calculated through the inverse calculation formula of deformation resistance. This actual deformation resistance value is then applied to the mechanism-data fusion leveling rolling force calculation model to establish a cross-process mechanism-data fusion leveling rolling force calculation model.

[0138] P = fB

[0139]

[0140] The relationship between each factor and the contact arc length determines the range of values ​​for each parameter to be determined, as shown in Table 5.

[0141] Table 5

[0142]

[0143]

[0144] Using real data from the leveling mill as a sample, and mapping the strip's entry material number to the cold rolling mill production data, the actual deformation resistance was calculated. 80% of the sample was selected for parameter training and optimization, while the remaining 20% ​​was used for prediction. The parameter optimization algorithm was a particle swarm optimization algorithm. Figure 11 As shown, the particle swarm optimization algorithm flow is as follows:

[0145] ④ Cross-process mechanism - Prediction results of the leveling rolling force calculation model based on data fusion

[0146] The predicted value of the leveling rolling force calculation model using cross-process mechanism-data fusion is as follows: Figure 12 As shown, the relative error between the predicted and actual values ​​is within 15% (94.02%), indicating that the established mechanism-data-based model for predicting rolling force across processes has high accuracy.

[0147] Example 2:

[0148] This invention also provides a cross-process leveling mill rolling force calculation device, comprising:

[0149] The building module is used to construct the cold continuous rolling process control model; the cold continuous rolling process control model includes: rolling force model, friction coefficient model and deformation resistance model;

[0150] The calculation module is used to obtain the deformation resistance dataset based on the cold continuous rolling process control model;

[0151] The prediction module is used to obtain a leveling rolling force calculation model based on cross-process data fusion from the deformation resistance dataset, and at the same time input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force.

[0152] As one embodiment of the present invention, the calculation module deforms the rolling force model, friction coefficient model and deformation resistance model by inverse operation based on the deformation resistance of cold rolling according to the actual production data of cold rolling, so as to obtain the average actual deformation resistance of each stand of cold rolling.

[0153] As one embodiment of the present invention, the prediction module includes:

[0154] The calculation unit is used to obtain the actual deformation resistance of the strip by inverse solving the average actual deformation resistance of each stand in the cold rolling mill;

[0155] The building unit is used to construct a leveling rolling force calculation model based on cross-process mechanism-data fusion, according to the average actual deformation resistance of each stand in cold continuous rolling and the actual deformation resistance of the strip.

[0156] The prediction unit is used to input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force.

[0157] Example 3:

[0158] This invention also provides a cross-process leveling mill rolling force calculation system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a cross-process leveling mill rolling force calculation method when run by the processor.

[0159] Example 4:

[0160] This invention also provides a storage medium storing a computer program that, when running, executes a method for calculating the rolling force of a leveling mill across multiple processes.

[0161] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for calculating the rolling force of a leveling mill that spans multiple processes, characterized in that, include: Step S1: Construct a cold continuous rolling process control model; wherein, the cold continuous rolling process control model includes: rolling force model, friction coefficient model and deformation resistance model; Step S2: Obtain the deformation resistance dataset based on the cold continuous rolling process control model; Step S3: Based on the deformation resistance dataset, obtain the leveling rolling force calculation model based on cross-process data fusion, and simultaneously input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force. In step S2, based on the actual production data of cold rolling mill, the rolling force model, friction coefficient model and deformation resistance model are deformed by inverse calculation based on the deformation resistance of cold rolling mill, so as to obtain the average actual deformation resistance of each stand of cold rolling mill. Step S3 includes: The actual deformation resistance of the strip is obtained by inverse solving the average actual deformation resistance of each stand in the cold continuous rolling mill. Based on the average actual deformation resistance of each stand in cold rolling mill and the actual deformation resistance of strip steel, a calculation model for leveling rolling force based on cross-process mechanism-data fusion is constructed. The actual production data of cold continuous rolling is input into the leveling rolling force calculation model to predict the rolling force.

2. A multi-process leveling mill rolling force calculation device, characterized in that, include: The building module is used to construct the cold continuous rolling process control model; the cold continuous rolling process control model includes: rolling force model, friction coefficient model and deformation resistance model; The calculation module is used to obtain the deformation resistance dataset based on the cold continuous rolling process control model; The prediction module is used to obtain a leveling rolling force calculation model based on cross-process data fusion according to the deformation resistance dataset, and at the same time input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force. The calculation module uses actual production data of cold rolling mill to deform the rolling force model, friction coefficient model and deformation resistance model through inverse operation based on the deformation resistance of cold rolling mill, so as to obtain the average actual deformation resistance of each stand of cold rolling mill. The prediction module includes: The calculation unit is used to obtain the actual deformation resistance of the strip by inverse solving the average actual deformation resistance of each stand in the cold rolling mill; The building unit is used to construct a leveling rolling force calculation model based on cross-process mechanism-data fusion, according to the average actual deformation resistance of each stand in cold continuous rolling and the actual deformation resistance of the strip. The prediction unit is used to input the actual production data of cold continuous rolling into the leveling rolling force calculation model to predict the rolling force.

3. A cross-process leveling mill rolling force calculation system, characterized in that, include: A memory and a processor, wherein the memory stores a computer program executed by the processor, the computer program executing the cross-process leveling mill rolling force calculation method as described in claim 1 when run by the processor.

4. A storage medium, characterized in that, The storage medium stores a computer program that, when running, executes the cross-process leveling mill rolling force calculation method as described in claim 1.

Images