Collapse deformation prediction method suitable for ultra-wide aluminum alloy intermediate billet in roller way transportation process

Abstract

A collapse deformation prediction method suitable for an ultra-wide aluminum alloy intermediate billet in the roller way transportation process comprises the steps that firstly, a finite element thermal coupling model in the aluminum alloy intermediate billet transportation process is established; secondly, finite element simulation is conducted on the aluminum alloy intermediate billets under different working conditions to obtain collapse deformation data, and a thermal coupling finite element model of the aluminum alloy intermediate billets is corrected through a field test; finally, working condition parameters including the width, the thickness, the initial temperature and the transportation time of the aluminum alloy intermediate billet serve as input data, the edge collapse deformation amount of the aluminum alloy intermediate billet in the transportation process serves as output data, and the input data and the output data form a training set; and constructing an aluminum alloy intermediate billet collapse deformation prediction model based on the improved support vector machine. The prediction method provided by the invention has the advantages of high accuracy, good generalization, reliable stability, excellent engineering practical comprehensive performance and wide application range.

Description

Technical Field

[0001] The present invention relates to the technical field of strip rolling, and particularly relates to a method for predicting the collapse deformation during the roller table transportation of ultra-wide aluminum alloy intermediate billets. Background Art

[0002] Aluminum alloy is an important basic material for the development of the national economy. Especially, high-performance and large-size aluminum alloy strips have become important structural materials in the fields of national defense, aviation, transportation, etc. Due to its good plasticity and low deformation resistance, aluminum alloy is easy to control during the rolling process, can maximize the equipment capacity, greatly reduce the energy consumption during the material preparation process, reduce environmental pollution, and create the greatest economic benefits while ensuring product quality and production efficiency. Hot continuous rolling of aluminum is one of the fastest developing and most widely used aluminum alloy production technologies. Due to the rolling characteristics such as large size, multiple specifications, and complex processes, the shape quality control of ultra-wide aluminum alloy strip hot continuous rolling has always been a research hotspot in the field of aluminum alloy rolling, and there are bottleneck problems in achieving stable and efficient production.

[0003] During the hot continuous rolling production process of aluminum alloy strips, the aluminum alloy intermediate billets are transported from the rough rolling outlet to the finish rolling inlet through the tapered roller table. During the transportation process, due to the influence of factors such as the yield strength of the aluminum alloy material under high-temperature conditions, the aluminum alloy intermediate billets undergo collapse deformation on the tapered transportation roller table, resulting in problems of adhesion and scratching due to the contact between the intermediate billets and the transportation roller table caused by the collapse deformation, seriously affecting its subsequent stable rolling production, being difficult to meet the market's requirements for the shape quality of ultra-wide aluminum alloy, and further restricting the efficient, low-cost and stable production of ultra-wide (more than 3m) aluminum alloy strips.

[0004] At present, relevant experts and scholars have conducted some research on the problem of predicting the collapse deformation of aluminum alloy billets, mainly including three methods: experimental method, mathematical model method and finite element method.

[0005] For example, Li Yuan of Baosteel Co., Ltd. analyzed the main reasons for the formation of transverse and longitudinal depressions on the slab surface in the article "Causes and Countermeasures for Surface Depressions of Slabs" (Baosteel Technology, No. 6, 2017, pp. 53-59), and proposed a series of effective measures to solve the slab depression, thereby improving the surface quality of the slab;

[0006] Li Zhaoxia of Tsinghua University studied the influence of process parameters on the width surface depression deformation of semi-continuous casting aluminum slabs in the article "Predicting the Depression Deformation of Semi-continuous Cast Aluminum Slabs Using Neural Networks" (Proceedings of the 2002 China Materials Symposium, Materials Preparation Process Simulation, pp. 1663-1667), and used the BP neural network to predict the width surface depression deformation of the slab. The results show that this prediction method can quickly and accurately predict the deformation amount of the slab width surface under given process conditions;

[0007] Lu Yongjian from Chongqing University analyzed the formation reasons and influencing factors of longitudinal depression on the wide face edge of continuous casting slabs in the article "Prevention of Longitudinal Depression on the Wide Face Edge of Continuous Casting Slabs" (Continuous Casting, November 2011, Issue 6, pp. 33 - 37). Combining with on-site production tests, a series of effective measures such as optimizing the mold flow field, optimizing the mold taper, and strengthening the cooling intensity of the narrow face in the mold foot roll section were proposed, effectively improving the longitudinal depression on the wide face edge of continuous casting slabs;

[0008] Yu Haijun from the University of Science and Technology Beijing established a dynamic simulation model for the coiling process at the exit of the cleaning line using ABAQUS finite element software in the doctoral thesis "Research on the Generation Mechanism and Control of Shape Defects in Wide-width Aluminum Foil Strip" (Doctoral Thesis of the Institute of Engineering Technology, University of Science and Technology Beijing, December 21, 2020). The influences of aluminum coil convexity, aluminum foil length-width ratio, aluminum foil thickness, coiling tension, aluminum foil yield strength, and initial shape defects of aluminum foil on the transverse wrinkling and buckling during the aluminum foil coiling process were studied, and the influence laws of various factors on the transverse wrinkling and buckling of aluminum foil were obtained, realizing the prediction of aluminum foil buckling, but lacking a prediction method for collapse during the transportation process of aluminum alloy.

[0009] From the above analysis, it can be seen that the current research on the collapse deformation problem of the intermediate billet mainly focuses on the casting process, and there is little research on the collapse deformation during the roller table transportation process of the intermediate billet in the publicly available materials. In particular, there is no research on the collapse deformation during the roller table transportation process of ultra-wide aluminum alloy intermediate billets that are difficult to meet market requirements. Summary of the Invention

[0010] The purpose of the present invention is to propose a method for predicting collapse deformation during the roller table transportation process of ultra-wide aluminum alloy intermediate billets, to solve the problem that the collapse deformation during the transportation process of ultra-wide aluminum alloy intermediate billets with a width of more than 3m is difficult to quantitatively predict and analyze, resulting in the inability to guarantee the shape quality.

[0011] To achieve the above purpose, the technical solution of the present invention is:

[0012] A method for predicting collapse deformation during the roller table transportation process of ultra-wide aluminum alloy intermediate billets, including establishing a thermo-mechanical coupling finite element model during the transportation process of aluminum alloy intermediate billets; obtaining collapse deformation data through finite element simulation using aluminum alloy intermediate billets under different working conditions; and finally constructing a prediction model for the collapse deformation of aluminum alloy intermediate billets based on an improved support vector machine.

[0013] Specifically, the method for predicting collapse deformation during the roller table transportation process of ultra-wide aluminum alloy intermediate billets described in the present invention includes the following steps:

[0014] 1) Establish a three-dimensional thermo-mechanical coupling finite element model for the roller table transportation of aluminum alloy intermediate billets, including establishing the physical models of aluminum alloy intermediate billets and transportation roller tables, importing the material constitutive relation model, creating boundary conditions, and mesh generation;

[0015] 2) Use the finite element analysis method to solve the established three-dimensional thermo-mechanical coupling finite element model for the roller table transportation of aluminum alloy intermediate billets, obtain the temperature field, stress field, and strain field of the aluminum alloy intermediate billets during the roller table transportation process, compare the measured temperature distribution of the aluminum alloy intermediate billets on-site with the simulation results of the finite element analysis method, and correct and verify the accuracy and precision of the corresponding three-dimensional thermo-mechanical coupling finite element model for the roller table transportation process of aluminum alloy intermediate billets;

[0016] 3) Combine the corrected three-dimensional thermo-mechanical coupling finite element model for the roller table transportation of aluminum alloy intermediate billets, simulate the collapse deformation phenomenon during the roller table transportation of aluminum alloy intermediate billets by setting different process parameters, reveal the law between the process parameters of the intermediate billets and the collapse deformation amount during the roller table transportation of aluminum alloy intermediate billets, and obtain the edge collapse deformation amount data under different working conditions during the transportation process of aluminum alloy intermediate billets; The different process parameters include but are not limited to different widths, thicknesses, initial temperatures, and transportation times of aluminum alloy intermediate billets;

[0017] 4) Take the edge collapse deformation simulation data during the roller table transportation of aluminum alloy intermediate billets as a data set, construct a collapse deformation prediction model for the roller table transportation of aluminum alloy intermediate billets based on an improved support vector machine, and use the prediction model to predict the collapse deformation phenomenon of aluminum alloy intermediate billets under different working conditions in actual production on-site.

[0018] Preferably, step 1) establishing a three-dimensional thermo-mechanical coupling finite element model for the roller table transportation of aluminum alloy intermediate billets includes:

[0019] ① According to the actual on-site process parameters of aluminum alloy intermediate billets, use Creo software to establish the physical model of the roller table transportation of aluminum alloy intermediate billets, and import the physical model into the finite element software ABAQUS for simulation analysis;

[0020] ② Define the material properties, set the material property parameters of aluminum alloy materials according to the thermal property parameters exported by JMatPro software, establish the UHARD user material subroutine, and import the material constitutive relation model into the finite element software ABAQUS for simulation analysis;

[0021] The material constitutive relation model is expressed as:

[0022]

[0023] In the formula, Z is the strain rate factor compensated by temperature,

[0024]

[0025] Wherein, σ is the true stress, in MPa; α is the stress level parameter, is the strain rate, in s -1 ; A and n are constants independent of temperature, Q is the deformation activation energy, in KJ·mol -1 ; R is the gas constant, 8.314 J·mol -1 ·℃ -1 ; T is the absolute temperature, in K;

[0026] ③ Analysis step and contact property setting; Define the contact condition between the aluminum alloy intermediate blank and the conveying roller table, set the intermediate blank as the master surface and the conical roller table as the slave surface;

[0027] ④ Boundary condition setting: Apply the gravity load boundary condition to the aluminum alloy intermediate blank, and set the temperature boundary conditions of heat convection, heat radiation and heat conduction for the aluminum alloy intermediate blank. Among them, the heat convection coefficient, in W / (m 2 ·℃), is 1.352 - 2.641; the radiation heat transfer coefficient, in W / (m 2 ·℃), is 6.556 - 13.817; the thermal conductivity, in W / (m·℃), is 169.3 - 180.3;

[0028] The initial temperature condition of the intermediate blank is set as the surface temperature equation of the intermediate blank measured by a thermal imager and fitted:

[0029] T t = 472.44 + 0.09x t - 2.05×10 -4 x t 2 + 1.65×10 -7 x t 3 - 4.31×10 -11 x t 4

[0030] Wherein, x t is the abscissa of the intermediate blank along the width direction, in mm, and T t is the temperature corresponding to this width coordinate, in ℃;

[0031] ⑤ Mesh generation, set the mesh type as the hexahedral element mesh of DC3D8;

[0032] Import the surface temperature equation of the aluminum alloy intermediate blank as a predefined field into the dynamic transportation model of the aluminum alloy intermediate blank to form a three-dimensional thermo-mechanical coupling finite element model during the transportation process of the aluminum alloy intermediate blank on the roller table, so as to conduct stress and deformation analysis of the aluminum alloy intermediate blank.

[0033] Preferably, for the prediction model of the collapse deformation during the transportation of the aluminum alloy intermediate billet based on the improved support vector machine in step 4), the genetic algorithm GA is used to optimize the kernel function parameters and penalty coefficients of the support vector machine, and the optimized kernel function parameters, penalty coefficients and the collapse deformation data obtained by simulation are used to construct the prediction model of the collapse deformation during the transportation of the aluminum alloy intermediate billet.

[0034] Preferably, the prediction method in step 4) includes:

[0035] a) Taking the working conditions including the initial temperature, width, thickness and transportation time of the intermediate billet as input data, taking the data including the edge collapse of the intermediate billet and the length of the contact line with the roller table as output data, combining the input data and output data as a data set, randomly sampling the data obtained from the simulation model, at least 60 groups of data, dividing them into a training set and a test set, where the amount of the training set is larger than that of the test set, and using the mapminmax function to normalize the data;

[0036] b) Setting the initial parameters of the improved support vector machine, calculating the corresponding fitness value, evaluating the fitness function, and using the genetic algorithm GA to optimize the kernel function parameters and penalty coefficients of the support vector machine;

[0037] c) Judging whether the termination condition is reached. If it is reached, the calculation ends and the optimal prediction value is output. If it is not reached, repeat step b);

[0038] d) Substituting the parameters c and g found by the genetic algorithm GA into the SVM model for retraining to obtain the optimal SVM model; using the svmpredict function for prediction and the mapminmax function to denormalize the results to obtain the prediction result of the collapse deformation during the transportation of the aluminum alloy intermediate billet.

[0039] Preferably, in step a), the ratio of the training set to the test set is ≥ 3:1.

[0040] In the prediction method of the present invention:

[0041] Step 1) Construct a three-dimensional thermo-mechanical coupling finite element model during the roller table transportation of the aluminum alloy intermediate billet, including establishing the geometric models of the aluminum alloy intermediate billet and the transportation roller table, importing the material constitutive relation model, creating assembly constraints and boundary conditions, and mesh generation;

[0042] Step 2) Use the finite element analysis method to solve the established three-dimensional thermo-mechanical coupling model during the roller table transportation of the aluminum alloy intermediate billet to obtain the temperature field, stress field and strain field during the roller table transportation of the aluminum alloy intermediate billet, compare the measured temperature distribution on site with the finite element simulation results, and correct and verify the accuracy of the corresponding three-dimensional thermo-mechanical coupling model during the roller table transportation of the aluminum alloy intermediate billet;

[0043] Step 3) Combine the three-dimensional thermo-mechanical coupling model during the roller table transportation of the modified aluminum alloy intermediate billet, simulate the collapse deformation phenomenon during the transportation of the aluminum alloy intermediate billet by setting different working conditions, reveal the law between the process parameters of the intermediate billet and the collapse deformation amount during the roller table transportation of the aluminum alloy intermediate billet, and obtain the edge collapse deformation amount data under different working conditions during the transportation of the aluminum alloy intermediate billet.

[0044] Step 4) Take the edge collapse deformation simulation data during the transportation of the aluminum alloy intermediate billet as the data set, construct a collapse deformation prediction model for the aluminum alloy intermediate billet during transportation based on the improved support vector machine, and use the prediction model to predict the collapse deformation amount of the aluminum alloy intermediate billet under different working conditions under the actual production conditions on site.

[0045] Furthermore, the material constitutive relation model is expressed as:

[0046]

[0047] In the formula, Z is the strain rate factor compensated by temperature, which can be expressed as:

[0048]

[0049] In the formula, σ is the true stress, MPa; α is the stress level parameter, is the strain rate, s -1 ; A, n are constants independent of temperature, Q is the deformation activation energy, KJ·mol -1 ; R is the gas constant, 8.314J·mol -1 ·℃ -1 ; T is the absolute temperature, K;

[0050] Furthermore, correct the three-dimensional thermo-mechanical coupling model during the roller table transportation of the aluminum alloy intermediate billet constructed in Step 1 through the on-site measured temperature distribution in Step 2 to improve the accuracy of the three-dimensional thermo-mechanical coupling simulation during the roller table transportation of the aluminum alloy intermediate billet.

[0051] Furthermore, the process parameters of the aluminum alloy intermediate billet in Step 3, namely the width, thickness, initial temperature, and transportation time of the aluminum alloy intermediate billet, are used to simulate and obtain the collapse deformation data by setting different process parameters of the intermediate billet during the transportation of the aluminum alloy intermediate billet.

[0052] Furthermore, for the collapse deformation prediction model of the aluminum alloy intermediate billet during transportation based on the improved support vector machine in Step 4, use the genetic algorithm to optimize the kernel function parameters and penalty coefficients of the support vector machine, and construct the collapse deformation prediction model for the aluminum alloy intermediate billet during transportation with the optimized kernel function parameters, penalty coefficients, and the collapse deformation data obtained by simulation.

[0053] Advantages or beneficial effects of the present invention:

[0054] 1) The present invention uses the collapse deformation data obtained by finite element model simulation during the transportation of the intermediate billet as the data set for improving the support vector machine, and constructs a prediction model for the collapse deformation during the transportation of the ultra-wide aluminum alloy intermediate billet. The results show that the error between the predicted value and the actual value is small, and the accuracy of the prediction model is high, which can effectively predict the collapse deformation during the roller path transportation of the aluminum alloy intermediate billet quickly and accurately.

[0055] 2) The collapse deformation prediction model proposed by the present invention during the transportation of the ultra-wide aluminum alloy intermediate billet can simulate and predict the intermediate billets of different specifications. The prediction model has a wide range of applications and can be used to guide the determination of the limit transportation size specifications of the intermediate billets of each grade of aluminum alloy during normal and stable rolling production, providing a practical and innovative implementation path for solving the shape quality problems caused by the high-temperature collapse deformation of the ultra-wide aluminum alloy intermediate billet.

[0056] 3) The present invention uses the proposed prediction method to obtain the collapse amount during the transportation of the aluminum alloy intermediate billet through off-line simulation prediction. The prediction method can not only obtain the accurate collapse deformation amount of the aluminum alloy under different process parameters, but also reduce the actual R & D costs and time costs of the enterprise. Description of the drawings

[0057] Figure 1 Schematic diagram of the thermo-mechanical coupling finite element model during the transportation of the aluminum alloy intermediate billet provided by the embodiment of the present invention;

[0058] Figure 2 Flow chart of the thermo-mechanical coupling finite element model simulation during the transportation of the aluminum alloy intermediate billet;

[0059] Figure 3 Schematic diagram of the comparison between the simulated temperature and the measured temperature of the aluminum alloy intermediate billet;

[0060] Figure 4 Flow chart of the prediction model based on the improved support vector machine;

[0061] Figure 5 Comparison diagram of the results of the prediction model based on the improved support vector machine; Detailed implementation manners

[0062] The present invention will be further described below in conjunction with the embodiments and the drawings.

[0063] The method for predicting the collapse deformation during the roller - table transportation of ultra - wide aluminum alloy intermediate billets according to the present invention includes establishing a thermo - mechanical coupling finite - element model during the transportation of aluminum alloy intermediate billets; obtaining collapse deformation data through finite - element simulation using aluminum alloy intermediate billets under different working conditions; and finally constructing a prediction model for the collapse deformation of aluminum alloy intermediate billets based on an improved support vector machine.

[0064] The specific steps are as follows:

[0065] Step 1) Construct a three - dimensional thermo - mechanical coupling finite - element model during the roller - table transportation of aluminum alloy intermediate billets. As Figure 1 shown, the specific establishment process is as Figure 2 shown, mainly including the following:

[0066] ① According to the actual process parameters of the aluminum alloy intermediate billet on - site, use Creo software to establish a physical model of the roller - table transportation of the aluminum alloy intermediate billet, and import the physical model into the large - scale finite - element software ABAQUS for simulation analysis;

[0067] ② Define the material properties. Set the material property parameters of the aluminum alloy according to the thermal physical property parameters exported by JMatPro software. As shown in Table 1, import the constitutive relation model into the simulation software through the established UHARD user material subroutine;

[0068] Table 1 Aluminum alloy material property parameters

[0069] Parameter Name Unit Value Thermal Conductivity W / (m·℃) 188.57 Heat Convection Coefficient <![CDATA[W / (m 2 ·°C)]]> 2.6407 Radiation Heat Transfer Coefficient <![CDATA[W / (m 2 ·°C)]]> 13.817 Density <![CDATA[kg / m 3 > 2720 Specific Heat Capacity J / kg·℃ 1110 Coefficient of Thermal Expansion 1 / ℃ / (1 / K) <![CDATA[27.28(10e -6 / K)]]> Young's Modulus Pa <![CDATA[50.05×10 9 > Acceleration of Gravity <![CDATA[m / s 2 > 9.8 Poisson's Ratio - 0.32 Bulk Modulus GPa 59.47 Shear Modulus GPa 18.4

[0070] The aluminum alloy constitutive relation model is expressed as:

[0071]

[0072] In the formula, Z is the strain - rate factor compensated for temperature, expressed as:

[0073]

[0074] In the formula, σ is the true stress, MPa; α is the stress - level parameter, is the strain rate, s -1 ; A, n are constants independent of temperature, Q is the deformation activation energy, KJ·mol -1 ; R is the gas constant, 8.314J·mol -1 ·℃ -1 ; T is the absolute temperature, K.

[0075] ③ Set the analysis steps and contact properties. On the basis of the initial analysis step, set the thermal analysis step for transient heat - transfer analysis. Define the contact conditions between the aluminum alloy intermediate billet and the transportation roller table, set the intermediate billet as the master surface and the conical roller table as the slave surface.

[0076] ④ Boundary condition setting: Apply a gravity load boundary condition to the aluminum alloy intermediate blank. Set the temperature boundary conditions of heat convection, heat radiation, and heat conduction for the aluminum alloy intermediate blank. The coefficients of heat convection, heat radiation, and heat conduction are shown in Table 1. The initial temperature condition of the intermediate blank is set as the surface temperature equation of the intermediate blank measured by a thermal imager and fitted:

[0077] T t = 472.44 + 0.09x t - 2.05×10 -4 x t 2 + 1.65×10 -7 x t 3 - 4.31×10 -11 x t 4

[0078] where x t is the abscissa of the intermediate blank along the width direction, and T t is the temperature corresponding to this width coordinate.

[0079] ⑤ Mesh generation: Set the mesh type as a hexahedral element mesh of DC3D8.

[0080] Import the temperature field of the aluminum alloy intermediate blank (the surface temperature equation of the intermediate blank) obtained from the simulation as a predefined field into the dynamic transportation simulation model of the aluminum alloy intermediate blank (simulation analysis using the finite element software ABAQUS) to form a three-dimensional thermal-mechanical coupling model during the roller conveyor transportation process of the aluminum alloy intermediate blank, so as to conduct stress and deformation analysis of the aluminum alloy intermediate blank.

[0081] Step 2) Use the finite element analysis method to solve the established three-dimensional thermal-mechanical coupling model during the roller conveyor transportation process of the aluminum alloy intermediate blank, obtain the temperature field, stress field, and strain field during the roller conveyor transportation process of the aluminum alloy intermediate blank, compare the measured temperature distribution on site with the finite element simulation results, as Figure 3 shown, correct (the surface temperature equation of the intermediate blank) and verify the accuracy of the corresponding three-dimensional thermal-mechanical coupling model during the roller conveyor transportation process of the aluminum alloy intermediate blank to ensure the reliability of the sample data obtained from the finite element simulation;

[0082] Step 3) Combine the three-dimensional thermo-mechanical coupling model during the roller table transportation of the modified aluminum alloy intermediate billet, and simulate the collapse deformation phenomenon during the transportation of the aluminum alloy intermediate billet by setting process parameters such as different widths, thicknesses, initial temperatures, and transportation times, reveal the laws between the width, thickness, initial temperature, and transportation time of the intermediate billet and the collapse deformation amount during the roller table transportation of the aluminum alloy intermediate billet, and obtain the edge collapse deformation amount data under different widths, thicknesses, initial temperatures, and transportation time conditions during the transportation of the aluminum alloy intermediate billet;

[0083] Step 4) Take the edge collapse deformation amount data obtained by simulation under different widths, thicknesses, initial temperatures, and transportation time conditions during the transportation of the aluminum alloy intermediate billet as a data set, and construct a collapse deformation prediction model for the aluminum alloy intermediate billet during transportation based on an improved support vector machine. The prediction process is as Figure 4 shown, and the specific steps are as follows:

[0084] a) Take the working conditions including the initial temperature, width, thickness, and transportation time of the intermediate billet as input data, take the data including the length of the contact line between the edge collapse of the intermediate billet and the roller table as output data, combine the input data and output data as a data set, randomly sample the data obtained from 68 groups of simulation models, where 75% is used as the training set and 25% is used as the test set, and use the mapminmax function to normalize the data;

[0085] b) Set the initial parameters of the improved support vector machine, calculate the corresponding fitness value, evaluate the fitness function, and use GA to optimize the kernel function parameters and penalty coefficients of the support vector machine;

[0086] c) Judge whether the termination condition is reached. If it is reached, the calculation ends and the optimal prediction value is output. If it is not reached, repeat step b);

[0087] d) Substitute the parameters c and g found by the GA algorithm into the SVM model for retraining to obtain the optimal SVM model; use the svmpredict function for prediction and the mapminmax function to denormalize the results to obtain the collapse deformation prediction results during the transportation of the aluminum alloy intermediate billet.

[0088] The prediction accuracy of the collapse deformation prediction model for the aluminum alloy intermediate billet during transportation established based on the improved support vector machine reaches 99.4% for the high-temperature collapse deformation during the transportation of the aluminum alloy intermediate billet. As Figure 5 shown, using this prediction model, the collapse deformation amount of the aluminum alloy intermediate billet under different working conditions under the actual production conditions on site can be predicted, providing theoretical and technical support for the stable production of aluminum alloy hot continuous rolling.

[0089] The rough rolling width of a certain aluminum industry's 1+3 hot tandem rolling mill is 4500 mm, and the finishing rolling width is 3300 mm. It can produce hot rough rolling products with a width of 4350 mm and hot finishing rolling products with a width of 3000 mm. The shape of the roller table from the rough rolling outlet to the finishing rolling inlet is a tapered roller. The aluminum alloy intermediate billet collapses and deforms on the tapered transport roller table, resulting in problems of adhesion and scratching due to the contact between the intermediate billet and the roller table during the collapse deformation, seriously affecting its subsequent stable rolling production and making it difficult to meet the market's requirements for the shape quality of ultra-wide aluminum alloy plates. Next, the application effects of the solution of this embodiment will be further described with application examples of aluminum alloy slabs of different grades.

[0090] Generally speaking, in the production site, it is stipulated that the edge of the aluminum alloy intermediate billet ≥ 100 mm contacts the tapered transport roller table (also known as the edge contact line length), then it is considered that the aluminum alloy intermediate billet has collapsed. Therefore, select the aluminum alloy intermediate billets of grades 3104, 3004, 5182, and 7075 under the limit specification dimensions produced on site, and use the prediction model of the present invention to predict the collapse deformation amount of the aluminum alloy intermediate billets under different grades of limit specifications. The collapse deformation prediction results of the aluminum alloy intermediate billets of different grades and size specifications are shown in Table 2.

[0091] As can be seen from Table 2, the prediction model of the collapse deformation during the transportation of the aluminum alloy intermediate billet established based on the improved support vector machine can accurately predict the collapse deformation amount of the aluminum alloy intermediate billets under different grades of limit specifications, and the prediction accuracy of whether the aluminum alloy intermediate billet collapses reaches 94.4%. On this basis, it can also predict the edge contact line length, and this prediction model can fully meet the production requirements of the production site for whether the aluminum alloy intermediate billet collapses.

[0092] Table 2 Collapse deformation prediction results of aluminum alloy intermediate billets of different grades and size specifications

[0093]

[0094]

[0095] In summary, a method for predicting the collapse deformation during the roller table transportation of ultra-wide aluminum alloy intermediate billets proposed by the present invention can accurately predict the collapse deformation amount during the roller table transportation of aluminum alloy intermediate billets, and can be used to guide the determination of the limit transportation size specifications during the normal and stable rolling production of aluminum alloy intermediate billets of each grade, reduce the actual R & D costs and time costs of enterprises, and provide a theoretical basis and implementation path for solving the shape quality problems caused by the high-temperature collapse deformation of aluminum alloy intermediate billets.

[0096] In addition, it should be noted that in this text, the terms "including", "comprising", or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or terminal device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article, or terminal device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article, or terminal device including the said element.

[0097] Finally, it should be noted that the above is the preferred embodiment of the present invention. It should be pointed out that although the preferred embodiments of the present invention have been described, for those skilled in the art of this technology, once the basic creative concept of the present invention is known, without departing from the principle described in the present invention, several improvements and refinements can still be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

Claims

1. A method for predicting the collapse deformation during the roller - table transportation of ultra - wide aluminum alloy intermediate billets, characterized in that, it includes the following steps: 1) Construct a three - dimensional thermo - mechanical coupling finite - element model during the roller - table transportation of aluminum alloy intermediate billets, including establishing the physical model of the aluminum alloy intermediate billet and the transportation roller table, importing the material constitutive relation model, creating boundary conditions, and mesh generation; 2) Use the finite - element analysis method to solve the three - dimensional thermo - mechanical coupling finite - element model established during the roller - table transportation of aluminum alloy intermediate billets, obtain the temperature field, stress field, and strain field of the aluminum alloy intermediate billet during the roller - table transportation of the aluminum alloy intermediate billet. Compare the measured temperature distribution of the aluminum alloy intermediate billet on - site with the simulation results of the finite - element analysis method, and correct and verify the accuracy and precision of the corresponding three - dimensional thermo - mechanical coupling finite - element model during the roller - table transportation of the aluminum alloy intermediate billet; 3) Combine the corrected three - dimensional thermo - mechanical coupling finite - element model during the roller - table transportation of the aluminum alloy intermediate billet, simulate the collapse deformation phenomenon during the roller - table transportation of the aluminum alloy intermediate billet by setting different process parameters, reveal the law between the intermediate - billet process parameters and the collapse deformation amount during the roller - table transportation of the aluminum alloy intermediate billet, and obtain the edge collapse deformation amount data under different working conditions during the transportation of the aluminum alloy intermediate billet; the different process parameters include but are not limited to different widths, thicknesses, initial temperatures, and transportation times of the aluminum alloy intermediate billet; 4) Use the edge collapse deformation simulation data during the roller - table transportation of the aluminum alloy intermediate billet as a data set to construct a collapse deformation prediction model for the roller - table transportation of the aluminum alloy intermediate billet based on an improved support vector machine, and use the prediction model to predict the collapse deformation phenomenon of the aluminum alloy intermediate billet under different working conditions in actual production on - site.

2. The method for predicting the collapse deformation during the roller - table transportation of ultra - wide aluminum alloy intermediate billets as described in claim 1, characterized in that, step 1) constructing the three - dimensional thermo - mechanical coupling finite - element model during the roller - table transportation of the aluminum alloy intermediate billet includes: ① According to the actual on - site process parameters of the aluminum alloy intermediate billet, use Creo software to establish the physical model of the roller - table transportation of the aluminum alloy intermediate billet, and import the physical model into the finite - element software ABAQUS for simulation analysis; ② Define the material properties, set the material property parameters of the aluminum alloy according to the thermal physical property parameters exported by JMatPro software, establish the UHARD user - defined material subroutine, and import the material constitutive relation model into the finite - element software ABAQUS for simulation analysis; The material constitutive relation model is expressed as: In the formula, Z is the temperature - compensated strain - rate factor, Wherein, σ is the true stress, in MPa; α is the stress level parameter, is the strain rate, in s -1 ; A and n are constants independent of temperature, Q is the deformation activation energy, in KJ·mol -1 ; R is the gas constant, 8.314 J·mol -1 ·°C -1 ; T is the absolute temperature, in K; ③ Set the analysis step and contact properties; define the contact conditions between the aluminum alloy intermediate billet and the transportation roller table, set the intermediate billet as the master surface, and the conical roller table as the slave surface; ④ The boundary conditions are set to apply a gravity load boundary condition to the aluminum alloy intermediate blank, and temperature boundary conditions of heat convection, heat radiation, and heat conduction are set for the aluminum alloy intermediate blank. Among them, the heat convection coefficient, unit W / (m 2 ·℃), is 1.352 - 2.641; the radiation heat transfer coefficient, unit W / (m 2 ·℃), is 6.556 - 13.817; the thermal conductivity, unit W / (m·℃), is 169.3 - 180.3; The initial temperature condition of the intermediate billet is set as the intermediate - billet surface temperature equation measured by a thermal imager and fitted: T t = 472.44 + 0.09x t - 2.05×10 -4 x t 2 + 1.65×10 -7 x t 3 - 4.31×10 -11 x t 4 where x t is the abscissa of the intermediate billet in the width direction, unit: mm, and T t is the temperature corresponding to this width coordinate, unit: °C; ⑤ Mesh generation, set the mesh type as a hexahedral element mesh of DC3D8. The surface temperature equation of the aluminum alloy intermediate billet is imported into the dynamic transportation model of the aluminum alloy intermediate billet as a predefined field to construct a three-dimensional thermo-mechanical coupling finite element model during the roller table transportation process of the aluminum alloy intermediate billet, so as to analyze the stress and deformation of the aluminum alloy intermediate billet.

3. The method for predicting the collapse deformation during the roller table transportation process of the ultra-wide aluminum alloy intermediate billet as described in claim 1, characterized in that for the collapse deformation prediction model of the aluminum alloy intermediate billet during the transportation process based on the improved support vector machine in step 4), the genetic algorithm GA is used to optimize the kernel function parameters and penalty coefficients of the support vector machine, and the optimized kernel function parameters, penalty coefficients and the collapse deformation data obtained by simulation are used to construct the collapse deformation prediction model during the transportation process of the aluminum alloy intermediate billet.

4. The method for predicting the collapse deformation during the roller table transportation process of the ultra-wide aluminum alloy intermediate billet as described in claim 1 or 3, characterized in that the prediction method in step 4) includes: a) Using the working conditions including the initial temperature, width, thickness and transportation time of the intermediate billet as input data, and the data including the edge collapse of the intermediate billet and the length of the contact line with the roller table as output data. Combining the input data and output data as a data set, randomly sampling the data obtained from the simulation model, at least 60 groups of data, and dividing them into a training set and a test set, where the amount of the training set is greater than that of the test set, and using the mapminmax function to normalize the data; b) Setting the initial parameters of the improved support vector machine, calculating the corresponding fitness value, evaluating the fitness function, and using the genetic algorithm GA to optimize the kernel function parameters and penalty coefficients of the support vector machine; c) Judging whether the termination condition is reached. If it is reached, the calculation ends and the optimal prediction value is output. If it is not reached, repeat step b); d) Substituting the parameters c and g found by the genetic algorithm GA into the SVM model for retraining to obtain the optimal SVM model; using the svmpredict function for prediction and the mapminmax function to denormalize the results to obtain the collapse deformation prediction result during the transportation process of the aluminum alloy intermediate billet.

5. The method for predicting the collapse deformation during the roller table transportation process of the ultra-wide aluminum alloy intermediate billet as described in claim 4, characterized in that in step a), the training set∶test set ≥ 3∶1.

Images