Hot rolling unit control method and device and medium

By using the target plate shape prediction model in the hot rolling mill, combining historical data and plate type, the convexity and flatness prediction accuracy during plate forming is improved, and the problem of insufficient plate forming control accuracy in the prior art is solved, and the improvement of the Product quality and stability.

CN119016519APending Publication Date: 2024-11-26BEIJING SHOUGANG AUTOMATION INFORMATION TECH

Patent Information

Application Number
CN202411082688.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

When the existing hot-rolled mill control model is formed, the convexity and straightness control accuracy is insufficient, resulting in poor product quality, especially when producing high-strength automotive boards and home appliance boards, it is easy to have inter-frame wave shape and local high points problems.

Method used

The target plate-shaped prediction model is adopted, and the sample set is trained, including historical plate parameters and equipment parameters, and the gradient enhancement decision tree model is used to predict the exit convexity and straightness of the rolled plate, and the plate type and equipment status are considered in the prediction model. Improve prediction accuracy.

Benefits of technology

It improves the accuracy of plate shape control of hot rolling mills, reduces the risk of wave shape between racks, and improves product quality and rolling stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119016519A_ABST
    Figure CN119016519A_ABST
Patent Text Reader

Abstract

The invention discloses a hot rolling unit control method and device and a medium. The method comprises the steps that target plate parameters and target equipment parameters serve as input, and target outlet convexity and target outlet straightness of a first to-be-rolled plate at an outlet of a hot rolling unit are predicted through a target plate shape prediction model; wherein the target plate shape prediction model is obtained by inputting a training sample set into a to-be-trained model for training, and the training sample set comprises historical plate parameters of a second to-be-rolled plate, historical equipment parameters of a hot rolling frame, and historical target outlet convexity and historical outlet straightness of the second to-be-rolled plate; the second to-be-rolled plate and the first to-be-rolled plate are of the same type. The prediction precision of the target outlet convexity and the target outlet straightness can be improved, and then the plate shape control effect of the hot rolling unit is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of hot rolling control technology, and in particular relates to a hot rolling unit control method, device and medium. Background Art

[0002] When the hot rolling mill performs hot continuous rolling on the plate, it is necessary to control the target parameters of the plate during forming according to the plate parameters of the plate itself and the equipment parameters of the hot rolling mill, such as controlling the convexity and straightness, to ensure the quality of plate forming.

[0003] In the related art, when the plate control model controls the target convexity and straightness of the plate, the parameter control has a problem of poor accuracy due to the unreasonable model design. Summary of the invention

[0004] The embodiments of the present application provide a hot rolling mill control method, device and medium, which can improve the prediction accuracy of the target outlet convexity and the target outlet flatness at least to a certain extent, thereby improving the plate shape control effect of the hot rolling mill.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0006] According to a first aspect of an embodiment of the present application, a hot rolling mill control method is provided, wherein the hot rolling mill comprises N hot rolling stands connected in sequence, where N is an integer greater than or equal to 1, and the method comprises:

[0007] Acquire target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand;

[0008] Taking the target plate parameters and the target equipment parameters as input, and using a target plate shape prediction model, predicting a target exit convexity and a target exit flatness of the first plate to be rolled at an exit of the hot rolling mill;

[0009] Among them, the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling mill, and historical target outlet convexity and historical target outlet straightness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled.

[0010] In some embodiments of the present application, based on the above scheme, before predicting the target exit convexity and target exit flatness of the first sheet to be rolled at the exit of the hot rolling mill using the target plate shape prediction model, the method further includes:

[0011] According to the first plate to be rolled and a preset corresponding relationship, a target plate shape prediction model is obtained in the prediction model set, wherein the preset corresponding relationship includes a corresponding relationship between each type of plate to be rolled and one of the plate shape prediction models.

[0012] In some embodiments of the present application, based on the above scheme, when N is greater than 1, taking the target plate parameters and the target equipment parameters as inputs and using the target plate shape prediction model to predict the target exit convexity and target exit flatness of the first plate to be rolled at the exit of the hot rolling mill group includes:

[0013] Taking the target plate parameters and the target equipment parameters as input, and using a target plate shape prediction model, predicting a target exit convexity of the first plate to be rolled at the exit of the hot rolling mill;

[0014] Starting from i=1, i∈N, for the i-th hot rolling stand, the target plate parameters and the target equipment parameters are used as input, and the target plate shape prediction model is used to predict the outlet flatness of the first plate to be rolled at the outlet of the i-th hot rolling stand;

[0015] Wherein, when i=1, the target plate parameters of the i-th hot rolling mill stand include: the entrance straightness of the first plate to be rolled at the entrance of the hot rolling mill group;

[0016] When i is greater than 1, the target plate parameters of the i-th hot rolling stand include: the exit straightness of the first plate to be rolled at the exit of the i-1-th hot rolling stand.

[0017] In some embodiments of the present application, based on the above scheme, after predicting the exit convexity and exit flatness of the first plate to be rolled at the exit of the i-th hot rolling stand using the target plate shape prediction model, the method further includes:

[0018] If the predicted outlet straightness of the first sheet to be rolled at the i-th hot rolling stand exceeds a flatness threshold, the flatness threshold is used as the outlet straightness of the second sheet to be rolled at the outlet of the i-th hot rolling stand.

[0019] In some embodiments of the present application, based on the above solution, before obtaining the target plate parameters of the first plate to be rolled and the target equipment parameters of the hot rolling stand, the method further includes:

[0020] Acquire a training sample set, wherein the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling stand, and historical target exit convexity and historical target exit straightness of the second plate to be rolled;

[0021] The training sample set is input into the model to be trained to obtain the target plate shape prediction model.

[0022] In some embodiments of the present application, based on the above solution, when obtaining the training sample set, the method further includes:

[0023] Acquire the initial target crown, actual crown and model-controlled crown of the second plate to be rolled at the exit of the hot rolling mill, wherein the model-controlled crown is obtained by the plate shape control model based on actual plate parameters and actual hot rolling parameters, the actual plate parameters are parameters of the second plate to be rolled in the process of forming the actual crown, and the actual hot rolling parameters are parameters of the hot rolling mill in the process of hot rolling the second hot rolled plate;

[0024] Obtaining a first difference between the initial target convexity and the actual convexity, and weighting the first difference using a smoothing coefficient to obtain a weighted convexity value;

[0025] A second difference between the model-controlled convexity and the weighted convexity value is obtained, and the second difference is used as the historical target export convexity.

[0026] In some embodiments of the present application, based on the above solution, after obtaining the first difference between the initial target convexity and the actual convexity, the method further includes:

[0027] Obtaining a first model-controlled flatness and a second model-controlled flatness, wherein the first model-controlled flatness is obtained by the plate shape control model based on actual plate parameters and actual hot rolling parameters, and the second model-controlled flatness is obtained by weighting the first model-controlled flatness based on a proportional convexity coefficient;

[0028] Acquire a third difference between the first model controlled flatness and the second model controlled flatness, and weight the third difference by using the first difference weighted based on the thickness proportional coefficient to obtain a weighted flatness value;

[0029] The historical target outlet straightness is obtained according to the weighted straightness value and the first model controlled straightness.

[0030] In some embodiments of the present application, based on the aforementioned scheme, the model to be trained includes a gradient boosting decision tree model.

[0031] According to a second aspect of an embodiment of the present application, a hot rolling mill control device is provided, wherein the hot rolling mill includes N hot rolling mill stands connected in sequence, where N is an integer greater than or equal to 1, and the device includes:

[0032] An acquisition unit, used for acquiring target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand;

[0033] A prediction unit, configured to take the target plate parameters and the target equipment parameters as inputs, and use a target plate shape prediction model to predict a target exit convexity and a target exit flatness of the first plate to be rolled at an exit of the hot rolling mill;

[0034] Among them, the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling mill, and historical target outlet convexity and historical target outlet straightness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled.

[0035] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by any method described in the first aspect.

[0036] According to a fourth aspect of an embodiment of the present application, an electronic device is provided, comprising one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the operations performed by any of the methods described in the first aspect.

[0037] One or more technical solutions provided by the embodiments of the present invention achieve at least the following technical effects or advantages:

[0038] In the present application, target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand are obtained; the target plate parameters and target equipment parameters are used as input, and a target plate shape prediction model is used to predict the target outlet convexity and target outlet flatness of the first plate to be rolled at the outlet of the hot rolling mill; wherein the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling stand, and historical target outlet convexity and historical target outlet flatness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled. Therefore, the present application trains the training model by utilizing the historical plate parameters of the second plate to be rolled of the same type as the first plate to be rolled, the historical equipment parameters of the hot rolling mill, and the historical target outlet convexity and historical target outlet flatness of the second plate to be rolled. The obtained target plate shape prediction model is a prediction model that matches the first plate to be rolled. Therefore, the target outlet convexity and target outlet flatness predicted by the target plate shape prediction model have higher accuracy, thereby making the control of the hot rolling unit more accurate and improving the plate shape control effect of the hot rolling unit.

[0039] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0041] Figure 1 A training principle diagram of a plate shape prediction model according to an embodiment of the present application is shown;

[0042] Figure 2 A flowchart of online application of the plate shape prediction model of an embodiment of the present application is shown;

[0043] Figure 3 A flow chart showing a hot rolling mill control method according to an embodiment of the present application is shown;

[0044] Figure 4 A structural diagram of a hot rolling mill control device according to an embodiment of the present application is shown;

[0045] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0047] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0048] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different models and / or processor devices and / or microcontroller devices.

[0049] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0050] It should also be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the objects used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those shown or described.

[0051] With the expansion of hot-rolled strip product varieties, changes in equipment status and users' increasing requirements for product quality, especially plate quality, plate quality problems and the resulting process problems are gradually exposed. Especially when producing high-strength automotive plates, home appliance plates and other products, problems such as wave shapes between frames and local high points are prone to occur, resulting in poor product quality.

[0052] During the hot rolling process, there are many factors affecting the shape control of the plate, such as rolling force, bending roll force, strip thickness, strip temperature, strip deformation resistance, initial roll shape of the roll, etc. The traditional shape control model uses many simplifications and assumptions, and the model considers fewer factors, which greatly reduces the accuracy of the model for shape control. Therefore, when the rolling state, steel type and specification change, the traditional shape control model can no longer adapt to the changing product requirements.

[0053] Based on this, an embodiment of the present application proposes a hot rolling mill control method, which can improve the prediction accuracy of the target outlet convexity and the target outlet flatness to a certain extent, thereby improving the plate shape control effect of the hot rolling mill.

[0054] The hot rolling mill control method according to the embodiment of the present application will be described below with reference to specific drawings.

[0055] See also Figure 1 , showing the training principle diagram of the plate shape prediction model of an embodiment of the present application.

[0056] like Figure 1 As shown, according to the first aspect of an embodiment of the present application, a hot rolling mill control method is provided, the hot rolling mill comprises N hot rolling mill frames connected in sequence, N is an integer greater than or equal to 1, and illustratively, the hot rolling mill of the embodiment of the present application comprises 6 hot rolling mill frames arranged in series, which are respectively defined as F1 to F6. For ease of understanding, the hot rolling mill control method is described below by taking F1 and F6 as examples.

[0057] The methods include but are not limited to:

[0058] Step S1. Acquire a training sample set, wherein the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling stand, and historical target outlet convexity and historical target outlet flatness of the second plate to be rolled;

[0059] Exemplarily, the main parameters of the hot rolling production line selected in the embodiment of the present application are: continuous rolling process section: six-stand four-roller tandem hot rolling unit; strip steel product thickness range: 1.5~19mm; strip steel product width range: 750~2130mm; strip steel products may include: automotive plates, pickled plates, welding bottle steel, pipeline steel, weathering steel, patterned plates, etc.

[0060] It can be understood that when determining the input characteristics of the model, the characteristic parameters that affect the convexity and flatness can be analyzed from a process perspective, or from a statistical perspective, the correlation analysis method can be used to analyze the correlation between various plate parameters, equipment parameters, etc. and convexity and flatness, and then determine the specific input parameters of the model.

[0061] For example, through the correlation analysis method, the correlation between one of the plate parameters and the convexity and straightness is shown in the following table:

[0062] Table 1 Correlation analysis values

[0063] Impact Factor Correlation value carbon 0.079 silicon 0.101 manganese 0.112 ... ... Target thickness 0.333 Target Width 0.272 Entrance straightness 0.425

[0064] Exemplarily, in an embodiment of the present application, historical plate parameters include but are not limited to: the content of each chemical element in the plate, target thickness, target width, target temperature, finishing entrance convexity, entrance flatness, intermediate billet thickness, etc.; wherein, finishing entrance convexity refers to the initial convexity of the second plate to be rolled at the entrance of the hot rolling mill, for example, 1 / 2 of the plate thickness; there are multiple entrance straightnesses, namely, the straightness of the second plate to be rolled at the entrance of each hot rolling stand, for example: the initial straightness of the second plate to be rolled at the entrance of the F1 stand is usually zero (because the plate does not meet the wave forming conditions).

[0065] Exemplarily, in the embodiments of the present application, the historical equipment parameters include but are not limited to: threading speed, rolling force of each stand, reduction rate of each stand, roll gap compensation coefficient of each stand, initial convexity of roll gap of each stand, number of rolling blocks of each stand, etc.

[0066] Based on the above, in the process of hot rolling mill rolling the plate to be rolled, various detection instruments installed in the rolling production line, including infrared pyrometers, pressure sensors, position sensors, etc., can be used to collect data on the above input parameters in order to construct a training sample set for the model.

[0067] In some embodiments, when obtaining the training sample set, the method further includes:

[0068] Step S11. Obtaining the initial target crown, actual crown and model-controlled crown of the second sheet to be rolled at the exit of the hot rolling mill, wherein the model-controlled crown is obtained by the plate shape control model based on actual sheet parameters and actual hot rolling parameters, the actual sheet parameters are parameters of the second sheet to be rolled in the process of forming the actual crown, and the actual hot rolling parameters are parameters of the hot rolling mill in the process of hot rolling the second hot rolled sheet;

[0069] It should be noted that the plate shape control model can refer to the existing plate shape control model in the relevant technology. When the plate shape control model performs plate shape control on the new plate to be rolled, it usually first obtains a theoretical target convexity, such as 20, based on the theoretical plate parameters of the plate to be rolled and the theoretical equipment parameters of the hot rolling unit. Since the calculation is based on the theoretical value, the calculated theoretical target convexity may deviate from the actual rolling effect. For example, after the second plate to be rolled is rolled by the hot rolling unit, the actual convexity may be 18. In addition, the initial target convexity may refer to the required convexity of the manufacturer or enterprise for the strip steel product, such as 15.

[0070] It can be understood that during the process of the hot rolling mill rolling the second plate to be rolled, the actual plate parameters and the actual equipment parameters can be detected, and then the actual plate parameters and the actual equipment parameters are used as inputs of the plate shape control model to correct the target convexity calculated by the plate shape control model, that is, to obtain the model-controlled convexity, for example, 20.

[0071] Step S12: obtaining a first difference between the initial target convexity and the actual convexity, and weighting the first difference with a smoothing coefficient to obtain a weighted convexity value;

[0072] Step S13. Obtain a second difference between the model-controlled convexity and the weighted convexity value, and use the second difference as the historical target export convexity.

[0073] For ease of understanding, the principles of the above steps S11 to S13 are exemplarily described using the following formula:

[0074] C New =C Cal -(C Ac -C target )*w

[0075] Where: C New is the value of convexity after self-learning, that is, the historical target export convexity, C Cal To control the convexity of the model, C Ac is the actual convexity, C target is the initial target convexity, and w is the smoothing coefficient.

[0076] For example: Based on the above formula, the convexity values ​​of the second sheet to be rolled are shown in the following table.

[0077] Table 2 Calculated convexity values

[0078] parameter Results / um <![CDATA[C New ]]> 37 <![CDATA[C Cal ]]> 44 <![CDATA[C Ac ]]> 52 <![CDATA[C target ]]> 40 w 0.6

[0079] Therefore, the embodiment of the present application further corrects the model control convexity through the initial target convexity and the actual convexity to obtain the historical target export convexity, thereby making the training samples more accurate and subsequently being able to improve the accuracy of model training.

[0080] In some embodiments, after obtaining a first difference between the initial target convexity and the actual convexity, the method further includes:

[0081] Step S14. Obtain the first model-controlled flatness and the second model-controlled flatness, wherein the first model-controlled flatness is obtained by the plate shape control model based on actual plate parameters and actual hot rolling parameters, and the second model-controlled flatness is obtained by weighting the first model-controlled flatness based on a proportional convexity coefficient;

[0082] Based on the above, it can be known that when the plate shape control model performs plate shape control on a new plate to be rolled, it usually first obtains a theoretical target flatness, such as 5, based on the theoretical plate parameters of the plate to be rolled and the theoretical equipment parameters of the hot rolling mill. Since the calculation is based on theoretical values, the calculated theoretical target flatness may deviate from the actual rolling effect. For example, after the second plate to be rolled is rolled by the hot rolling mill, the actual flatness may be 8. Then, by detecting the actual plate parameters and actual equipment parameters during the rolling of the second plate to be rolled by the hot rolling mill, and using the actual plate parameters and actual equipment parameters as the input of the plate shape control model, the target flatness calculated by the plate shape control model is corrected, that is, the first model-controlled flatness, such as 6, is obtained.

[0083] In addition, when the proportional convexity of the second plate to be rolled changes at the outlet of the hot rolling mill, due to the coupling between convexity and flatness, the outlet straightness will also change accordingly. Therefore, the proportional convexity coefficient is used to weight the control degree of the first model to obtain the second model controlled flatness, where the proportional convexity refers to: the thickness at the center minus the average thickness of the ends on both sides in the width direction of the plate, and the proportional convexity coefficient can be: 0.001.

[0084] Therefore, when establishing the training samples of the target outlet straightness, the coupling relationship between convexity and straightness is taken into account, so that the target outlet straightness is data related to the target outlet convexity, further improving the accuracy of subsequent model training.

[0085] Step S15. Obtain a third difference between the first model control flatness and the second model control flatness, and weight the third difference using the first difference weighted based on the thickness proportionality coefficient to obtain a weighted flatness value;

[0086] The thickness proportionality coefficient refers to: the ratio between the conversion relationship value between the outlet convexity and the outlet straightness and the outlet plate thickness. For example, the conversion relationship value is 1000 (corresponding to the above 0.001), and the outlet plate thickness is h.

[0087] Step S16: Obtain the historical target outlet straightness according to the weighted straightness value and the first model control straightness.

[0088] For ease of understanding, the principles of the above steps S14 to S16 are exemplarily described using the following formula:

[0089] f New,i =(f Cal2,i -f Cal1,i )*(-1000*(C Ac -C target ) / h)+f Cal1,i

[0090] Among them, f New,i is the value of the exit straightness of the ith 4 hot rolling mill stands after self-learning; f Cal1,i Controls the straightness for the first model; f Cal2,i The second model controls the straightness; h is the plate outlet thickness.

[0091] From the above, it can be seen that the flatness of the second plate at the exit of each hot rolling stand can be calculated by the above formula, and the flatness at the exit of the last hot rolling stand can be used as the historical target exit flatness.

[0092] For example, taking the F1 stand as an example, the various flatness values ​​of the second plate to be rolled are shown in the following table.

[0093] Table 3 Calculated straightness values

[0094] parameter result <![CDATA[f New,1 ]]> -221.9 <![CDATA[f Cal1,1 ]]> -184 <![CDATA[f Cal2,1 ]]> -105 h 2.5

[0095] Step S2: input the training sample set into the model to be trained to obtain the target plate shape prediction model.

[0096] In some embodiments, the model to be trained includes a gradient boosting decision tree model (GBDT); of course, the embodiments of the present application may also use other machine learning models for training, which will not be repeated here.

[0097] After obtaining the training sample set (i.e., the training sample set obtained after data preprocessing of the data obtained from historical production data collection), the training sample set is divided into a test set and a training set, and GBDT is trained and optimized. The optimization process includes but is not limited to: Gaussian process regression, point selection of acquisition function, whether the target is met, whether the number of iterations is met, and obtaining the optimal parameter combination.

[0098] It can be understood that for different types of strip steel products, the strip steel plates used may be different, and the corresponding plate parameters and equipment parameters may be different. Therefore, for each type of strip steel plate, the embodiment of the present application can be trained separately to obtain the corresponding target plate shape prediction model, thereby forming a prediction model set, which includes several plate shape prediction models, each plate shape prediction model corresponds to a strip steel plate, so that the corresponding plate shape prediction model can be used to perform parameter prediction for different types of strip steel plates in order to adapt to the diverse changes of strip steel products and improve the control accuracy of the hot rolling unit.

[0099] In addition, it is understandable that if the straightness value at the outlet of each rack is not limited in range, wavy shapes may appear after exceeding the threshold range. Therefore, when training the target plate shape prediction model, if the output straightness value exceeds the threshold range, the upper or lower limit of the range is used as the straightness value to reduce the risk of wavy shapes.

[0100] For example: The upper and lower limits of the outlet straightness of each hot rolling mill stand are as follows:

[0101] F1 F2 F3 F4 F5 F6 Upper limit 500 500 800 800 1000 1500 Lower limit -500 -500 -800 -800 -1000 -1500

[0102] See also Figure 2 , shows a flow chart of online application of the plate shape prediction model of an embodiment of the present application.

[0103] like Figure 2 As shown, thus, after step S1 to step S2, the target plate shape prediction model is trained. It can be understood that the training of the target plate shape prediction model can be deployed on a cloud server to reduce the load of the local server, or it can be deployed on a local server, without limitation here. In the subsequent application process, the target plate shape prediction model (plate shape multi-objective calculation model) can be automatically converted and deployed as a configuration file in a local server, such as an L2 server, so that the L2 server can read production data, use the model to perform parameter prediction and automatically update the model.

[0104] See also Figure 3 , showing a flow chart of a hot rolling mill control method according to an embodiment of the present application.

[0105] Step S3. Obtaining target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand;

[0106] It is understandable that the first plate to be rolled and the second plate to be rolled are the same type of plates. Therefore, when the plate prediction model matching the first plate to be rolled is used to predict parameters in the embodiment of the present application, the target outlet convexity and target outlet flatness obtained have higher accuracy, thereby making the control of the hot rolling mill more accurate and improving the plate shape control effect of the hot rolling mill.

[0107] Step S4. Taking the target plate parameters and the target equipment parameters as input, using the target plate shape prediction model, predicting the target exit convexity and target exit flatness of the first plate to be rolled at the exit of the hot rolling mill;

[0108] Among them, the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling mill, and historical target outlet convexity and historical target outlet straightness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled.

[0109] In some embodiments, before predicting the target exit convexity and target exit flatness of the first sheet to be rolled at the exit of the hot rolling mill using the target plate shape prediction model, the method further includes:

[0110] According to the first plate to be rolled and a preset corresponding relationship, a target plate shape prediction model is obtained in the prediction model set, wherein the preset corresponding relationship includes a corresponding relationship between each type of plate to be rolled and one of the plate shape prediction models.

[0111] Therefore, for different types of strip steel plates, corresponding plate shape prediction models can be used to perform parameter prediction to adapt to the diverse changes in strip steel products and improve the control accuracy of the hot rolling mill.

[0112] In some embodiments, when N is greater than 1, the target plate parameters and the target equipment parameters are used as inputs, and a target plate shape prediction model is used to predict a target exit convexity and a target exit flatness of the first plate to be rolled at the exit of the hot rolling mill, including:

[0113] Step S41. Taking the target plate parameters and the target equipment parameters as input, and using the target plate shape prediction model, predict the target exit convexity of the first plate to be rolled at the exit of the hot rolling mill;

[0114] Step S42. Starting from i=1, i∈N, for the i-th hot rolling stand, the target plate parameters and the target equipment parameters are used as inputs, and the target plate shape prediction model is used to predict the outlet flatness of the first plate to be rolled at the outlet of the i-th hot rolling stand;

[0115] Wherein, when i=1, the target plate parameters of the i-th hot rolling mill stand include: the entrance straightness of the first plate to be rolled at the entrance of the hot rolling mill group;

[0116] When i is greater than 1, the target plate parameters of the i-th hot rolling stand include: the exit straightness of the first plate to be rolled at the exit of the i-1-th hot rolling stand.

[0117] Specifically, using the target plate shape prediction model, the target exit convexity at the exit of the hot rolling mill is first predicted, and then the flatness at the exit of each stand is predicted in the order of the hot rolling stands, i.e., F1 to F6. For example, for the F1 stand, the entrance flatness (usually 0) at the entrance of the hot rolling mill is used as the model input to predict the exit flatness (P1) of the first plate to be rolled at the exit of the F1 stand. P1 is input into the model as the entrance flatness of the first plate at the entrance of the F2 stand, and the exit flatness (P2) of the first plate to be rolled at the exit of the F2 stand is predicted. By analogy, the exit straightness (P6) of the first plate to be rolled at the exit of the F6 stand is obtained. It can be understood that P6 is the target exit straightness of the first plate to be rolled at the exit of the hot rolling mill. Therefore, the exit straightness at the exit of the previous rack can be used as the entrance straightness at the entrance of the next rack, and the exit straightness of the first plate at the exit of each rack can be predicted in turn, so as to realize the allocation of hot rolling process parameters for the exit straightness of each rack. The straightness at the exit of the next rack is predicted by the straightness at the exit of the previous rack as the model input, so the exit straightnesses are related to each other, so that the exit straightness at the exit of the hot rolling unit obtained at the end is more accurate.

[0118] In some embodiments, after predicting the exit convexity and exit flatness of the first plate to be rolled at the exit of the i-th hot rolling stand using the target plate shape prediction model, the method further includes:

[0119] If the predicted outlet straightness of the first sheet to be rolled at the i-th hot rolling stand exceeds a flatness threshold, the flatness threshold is used as the outlet straightness of the second sheet to be rolled at the outlet of the i-th hot rolling stand.

[0120] It is understandable that, if the straightness value of each rack outlet is not limited in range, it may appear wavy after exceeding the threshold range. Therefore, if the predicted straightness value exceeds the threshold range, the upper limit or lower limit of the range is used as the straightness value to reduce the risk of wavy.

[0121] Therefore, in order to solve the problems of poor model calculation accuracy and complex production line conditions, the embodiment of the present application establishes a plate shape prediction model through the GBDT algorithm, so that it can accurately predict the convexity and straightness, thereby meeting the plate shape control requirements under different working conditions of different steel grades, making the convexity and straightness distribution of each frame more reasonable, reducing the risk of wave shape between frames, improving the cross-sectional profile of the strip, and improving product quality and rolling stability.

[0122] See also Figure 4 , showing a structural diagram of a hot rolling mill control device according to an embodiment of the present application.

[0123] like Figure 4 As shown, according to the second aspect of the embodiment of the present application, a hot rolling mill control device 200 is provided, wherein the hot rolling mill comprises N hot rolling mill frames connected in sequence, where N is an integer greater than or equal to 1, and the device 200 comprises:

[0124] An acquisition unit 201 is used to acquire target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand;

[0125] A prediction unit 202 is used to take the target plate parameters and the target equipment parameters as inputs, and use a target plate shape prediction model to predict a target exit convexity and a target exit flatness of the first plate to be rolled at an exit of the hot rolling mill;

[0126] Among them, the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling mill, and historical target outlet convexity and historical target outlet straightness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled.

[0127] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by any method described in the first aspect.

[0128] The computer readable storage medium may be a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the computer readable storage medium of the present application is not limited thereto, and in the present application, the readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus, or device.

[0129] The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more conductors, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0130] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0131] According to a fourth aspect of an embodiment of the present application, an electronic device is provided, comprising one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the operations performed by any of the methods described in the first aspect.

[0132] See also Figure 5 , which is a schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application.

[0133] According to the sixth aspect of an embodiment of the present application, an electronic device is provided, comprising one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement operations performed by any method of the first aspect.

[0134] like Figure 5 As shown, the electronic device 400 is in the form of a general computing device. The components of the electronic device 400 may include but are not limited to: at least one processing unit 410, at least one storage unit 420, and a bus 430 connecting different system components (including the storage unit 420 and the processing unit 410).

[0135] The storage unit stores program codes, which can be executed by the processing unit 410, so that the processing unit 410 executes the steps described in the above “Embodiment Method” section of this specification according to various exemplary implementations of the present application.

[0136] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 421 and / or a cache memory unit 422 , and may further include a read-only memory unit (ROM) 423 .

[0137] The storage unit 420 may also include a program / utility 424 having a set (at least one) of program modules 425, such program modules 425 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0138] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0139] The electronic device 400 may also communicate with one or more external devices 500 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 400, and / or communicate with any device that enables the electronic device 400 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 450. Furthermore, the electronic device 400 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 460. As shown, the network adapter 460 communicates with other modules of the electronic device 400 via a bus 430. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RA ID systems, tape drives, and data backup storage systems.

[0140] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of the present invention and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these. In addition, each functional unit may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit.

[0141] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0142] The units described as separate components may or may not be physically separated, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed in multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0143] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0144] The above description is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A hot rolling mill control method, characterized in that: The hot rolling mill group includes N hot rolling mill stands connected in sequence, where N is an integer greater than or equal to 1, and the method includes: Acquire target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand; Taking the target plate parameters and the target equipment parameters as input, and using a target plate shape prediction model, predicting a target exit convexity and a target exit flatness of the first plate to be rolled at an exit of the hot rolling mill; Among them, the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling mill, and historical target outlet convexity and historical target outlet straightness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled.

2. The method according to claim 1, characterized in that Before predicting the target exit convexity and target exit flatness of the first plate to be rolled at the exit of the hot rolling mill using the target plate shape prediction model, the method further includes: According to the first plate to be rolled and a preset corresponding relationship, a target plate shape prediction model is obtained in the prediction model set, wherein the preset corresponding relationship includes a corresponding relationship between each type of plate to be rolled and one of the plate shape prediction models.

3. The method according to claim 1, characterized in that: When N is greater than 1, the target plate parameters and the target equipment parameters are used as inputs, and a target plate shape prediction model is used to predict a target exit convexity and a target exit flatness of the first plate to be rolled at the exit of the hot rolling mill, including: Taking the target plate parameters and the target equipment parameters as input, and using a target plate shape prediction model, predicting a target exit convexity of the first plate to be rolled at the exit of the hot rolling mill; Starting from i=1, i∈N, for the i-th hot rolling stand, the target plate parameters and the target equipment parameters are used as input, and the target plate shape prediction model is used to predict the outlet flatness of the first plate to be rolled at the outlet of the i-th hot rolling stand; Wherein, when i=1, the target plate parameters of the i-th hot rolling mill stand include: the entrance straightness of the first plate to be rolled at the entrance of the hot rolling mill group; When i is greater than 1, the target plate parameters of the i-th hot rolling stand include: the exit straightness of the first plate to be rolled at the exit of the i-1-th hot rolling stand.

4. The method according to claim 3, characterized in that After predicting the outlet straightness of the first sheet to be rolled at the outlet of the i-th hot rolling stand by using the target plate shape prediction model, the method further includes: If the predicted outlet straightness of the first sheet to be rolled at the i-th hot rolling stand exceeds a flatness threshold, the flatness threshold is used as the outlet straightness of the second sheet to be rolled at the outlet of the i-th hot rolling stand.

5. The method according to claim 1, characterized in that Before obtaining the target plate parameters of the first plate to be rolled and the target equipment parameters of the hot rolling stand, the method further includes: Acquire a training sample set, wherein the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling stand, and historical target exit convexity and historical target exit straightness of the second plate to be rolled; The training sample set is input into the model to be trained to obtain the target plate shape prediction model.

6. The method according to claim 5, characterized in that When obtaining the training sample set, the method further includes: Acquire the initial target crown, actual crown and model-controlled crown of the second plate to be rolled at the exit of the hot rolling mill, wherein the model-controlled crown is obtained by the plate shape control model based on actual plate parameters and actual hot rolling parameters, the actual plate parameters are parameters of the second plate to be rolled in the process of forming the actual crown, and the actual hot rolling parameters are parameters of the hot rolling mill in the process of hot rolling the second hot rolled plate; Obtaining a first difference between the initial target convexity and the actual convexity, and weighting the first difference using a smoothing coefficient to obtain a weighted convexity value; A second difference between the model-controlled convexity and the weighted convexity value is obtained, and the second difference is used as the historical target export convexity.

7. The method according to claim 6, characterized in that After obtaining a first difference between the initial target convexity and the actual convexity, the method further includes: Obtaining a first model-controlled flatness and a second model-controlled flatness, wherein the first model-controlled flatness is obtained by the plate shape control model based on actual plate parameters and actual hot rolling parameters, and the second model-controlled flatness is obtained by weighting the first model-controlled flatness based on a proportional convexity coefficient; Acquire a third difference between the first model controlled flatness and the second model controlled flatness, and weight the third difference by using the first difference weighted based on the thickness proportional coefficient to obtain a weighted flatness value; The historical target outlet straightness is obtained according to the weighted straightness value and the first model controlled straightness.

8. The method according to claim 1, characterized in that: The model to be trained includes a gradient boosting decision tree model.

9. A hot rolling mill control device, characterized in that: The hot rolling mill group includes N hot rolling mill stands connected in sequence, where N is an integer greater than or equal to 1, and the device includes: An acquisition unit, used for acquiring target plate parameters of the first plate to be rolled and target equipment parameters of the hot rolling stand; A prediction unit, configured to take the target plate parameters and the target equipment parameters as inputs, and use a target plate shape prediction model to predict a target exit convexity and a target exit flatness of the first plate to be rolled at an exit of the hot rolling mill; Among them, the target plate shape prediction model is obtained by inputting a training sample set into the model to be trained, and the training sample set includes historical plate parameters of the second plate to be rolled, historical equipment parameters of the hot rolling mill, and historical target outlet convexity and historical target outlet straightness of the second plate to be rolled, and the second plate to be rolled is the same type of plate as the first plate to be rolled.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer program instruction, and the at least one computer program instruction is loaded and executed by a processor to implement the operations performed by any one of the methods according to claims 1-8.

Citation Information

Patent Citations

  • Method for controlling planeness of cold-rolling strip steel

    CN101618402A

  • Self-adaptation convexity change hot rolled plate shape control model

    CN103341503A

  • Hot continuous rolling strip steel convexity prediction method based on gradient boosting tree model

    CN110929347A

  • Dynamic cold continuous rolling plate shape prediction method based on deep recurrent neural network

    CN115007658A

  • Hot continuous rolling plate straightness classification prediction method based on deep belief network model

    CN117668667A

Cited By

  • Parameter determining method and apparatus, device, medium, and program product

    WO2026118407A1