Method and equipment for constructing roll bending force and roll shifting quantity budgeting model of cold rolling temper mill

Through deep learning technology, the bending force and rolling volume budget model of cold rolling flattening machine has been constructed, which solves the problem of insufficient adaptability and intelligence of pre-calculation of bending force in the existing technology, and achieves more accurate prediction and higher production efficiency.

CN118940005APending Publication Date: 2024-11-12SHOUGANG JINGTANG IRON & STEEL CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411035963.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing pre-calculation technology for bending roll force and rolling volume of cold rolling machine is severely lagging in terms of adaptability and intelligence, and cannot effectively deal with the differences in bending roll force demand caused by changes in different steel types and working conditions.

Method used

Using a budget model construction method based on deep learning, the network structure is determined by obtaining the roller count of the leveling machine, and the feature quantity data is input for training. The trained model is used to pre-calculate the bending force and the rolling amount, and the prediction is made. When the accuracy is low, the model parameters are updated through incremental self-learning method.

Benefits of technology

It improves the prediction accuracy of bending roller force and rolling volume, improves production efficiency and product quality, extends equipment life, and brings economic benefits to the enterprise.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940005A_ABST
    Figure CN118940005A_ABST
Patent Text Reader

Abstract

The invention provides a method and equipment for constructing a roll bending force and roll shifting quantity budgeting model of a cold rolling temper mill, and relates to the technical field of industrial big data analysis and deep learning, and the method comprises the steps: obtaining the number of rolls of the temper mill based on a production line; determining a network structure of the budgeting model based on the roller number; inputting the feature quantity source data into a budget model containing a network structure to train the budget model; and pre-calculating the roll bending force and the roll shifting amount by using the trained budgeting model. Therefore, the budgeting model can be fused with the rolling process technology of the cold rolling temper mill, product characteristics, deep learning, a neural network, model training and other technologies in multiple key fields, and accurate budgeting of the roll bending force and the roll shifting amount is achieved. The roll bending force and roll shifting amount set value prediction precision, the production efficiency, the product quality and the equipment service life can be improved, and remarkable economic benefits are brought to enterprises.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of industrial big data analysis and deep learning technology, and more specifically, to a method for constructing a cold rolling mill bending roll force and roll shifting amount budget model, a device for constructing a cold rolling mill bending roll force and roll shifting amount budget model, an electronic device and a storage medium. Background Art

[0002] The control targets of the same type of steel are very different. Moreover, with the changes in working conditions and customer needs, the control focus of the same product is different, which leads to different requirements for its bending roll force.

[0003] The existing control schemes do not effectively solve these problems, which leads to the serious lag in the adaptability and intelligence level of the pre-calculation of the bending roll force of the leveling machine.

[0004] Therefore, a new technical solution is urgently needed to solve the above technical problems. Summary of the invention

[0005] A series of simplified concepts are introduced in the Summary of the Invention, which will be further described in detail in the Detailed Description of the Invention. The Summary of the Invention does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0006] In a first aspect, the present invention proposes a method for constructing a cold rolling skin-pass mill bending force and roll shifting amount budget model, comprising:

[0007] Get the number of rollers of the skin-pass mill based on the production line;

[0008] Determine the network structure of the budget model based on the number of rollers;

[0009] Inputting the feature quantity source data into the budget model including the network structure to train the budget model;

[0010] The trained budget model is used to pre-calculate the bending roll force and roll shifting amount.

[0011] Optionally, the method further comprises:

[0012] The trained budget model is evaluated to update parameters of the budget model using an incremental self-learning method when the evaluation result indicates low prediction accuracy.

[0013] Optionally, when the evaluation result indicates that the prediction accuracy is low, an incremental self-learning method is used to update the parameters of the budget model, including:

[0014] Collect relevant data of products with low accuracy;

[0015] The relevant data are gradually input into the budget model to update the parameters of the budget model.

[0016] Optionally, before gradually inputting the relevant data into the budget model, the method further comprises:

[0017] Preprocess relevant data.

[0018] Optionally, the method further comprises:

[0019] Adjust the network structure of the budget model and / or increase training data based on the evaluation results.

[0020] Optionally, inputting the feature quantity source data into a budget model including a network structure to train the budget model includes:

[0021] The budget model parameters are optimized through the back-propagation algorithm.

[0022] Optionally, training the budget model further includes:

[0023] Regularization techniques are used during the training of the budget model.

[0024] Secondly, a device for constructing a cold rolling mill bending roll force and roll shifting amount budget model is also proposed, including:

[0025] Roller number determination module, used to obtain the number of rollers of the leveling machine based on the production line;

[0026] A structure determination module, used for determining a network structure of a budget model based on the number of rollers;

[0027] A training module, used for inputting the feature quantity source data into the budget model including the network structure to train the budget model;

[0028] The calculation module is used to pre-calculate the bending roll force and the roll shifting amount by using the trained budget model.

[0029] On the third aspect, an electronic device is also proposed, including a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions are used by the processor to execute the method for constructing the cold rolling mill bending roll force and roll shifting amount budget model as described above when the processor is running.

[0030] In a fourth aspect, a storage medium is also proposed, on which program instructions are stored, and the program instructions are used to execute the method for constructing the cold rolling mill bending roll force and roll shifting amount budget model as described above during operation.

[0031] According to the above technical solution, the number of rollers of the leveling mill is obtained based on the production line; the network structure of the budget model is determined based on the number of rollers; the characteristic source data is input into the budget model containing the network structure to train the budget model; and the trained budget model is used to pre-calculate the bending roll force and the amount of roll shifting. Therefore, the budget model can be used to integrate the rolling process technology, product characteristics, deep learning, neural network, model training and other key fields of the cold rolling leveling mill to achieve accurate budgeting of the bending roll force and the amount of roll shifting, which will help improve the prediction accuracy of the setting values ​​of the bending roll force and the amount of roll shifting, production efficiency, product quality and equipment life, and bring significant economic benefits to the enterprise.

[0032] The method for constructing the cold rolling mill bending roll force and roll shifting amount budget model of the present invention, and other advantages, objectives and features of the present invention will be reflected in part through the following description, and will also be understood by technical personnel in this field through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0034] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0035] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the exemplary embodiments below. The accompanying drawings are only for the purpose of illustrating exemplary embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0036] Figure 1 A schematic flow chart showing a method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to an embodiment of the present invention;

[0037] Figure 2 A schematic diagram showing a network structure of a budget model according to an embodiment of the present invention;

[0038] Figure 3 A schematic diagram showing predicted values ​​and actual values ​​of roll bending force and roll shifting amount according to an embodiment of the present invention;

[0039] Figure 4 A schematic diagram of a loss function according to an embodiment of the present invention is shown;

[0040] Figure 5 A schematic flow chart of an incremental self-learning process of calling a leveling machine roll bending force and roll shifting amount budget model according to an embodiment of the present invention is shown;

[0041] Figure 6 A schematic block diagram showing a device for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to an embodiment of the present invention; and

[0042] Figure 7 A schematic block diagram of an electronic device according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0043] According to the above technical solution, the number of rollers of the leveling mill is obtained based on the production line; the network structure of the budget model is determined based on the number of rollers; the characteristic source data is input into the budget model containing the network structure to train the budget model; and the trained budget model is used to pre-calculate the bending roll force and the amount of roll shifting. Therefore, the budget model can be used to integrate the rolling process technology, product characteristics, deep learning, neural network, model training and other key fields of the cold rolling leveling mill to achieve accurate budgeting of the bending roll force and the amount of roll shifting, which will help improve the prediction accuracy of the setting values ​​of the bending roll force and the amount of roll shifting, production efficiency, product quality and equipment life, and bring significant economic benefits to the enterprise.

[0044] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0045] The terms "first", "second", "third", "fourth", etc. (if any) 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 data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.

[0047] In order to solve the above technical problems, according to the first aspect of the present application, a method for constructing a cold rolling mill bending roll force and roll shifting amount budget model is proposed. Figure 1 FIG. 1 is a schematic flow chart of a method 100 for constructing a cold rolling mill bending force and roll shifting amount estimation model according to an embodiment of the present invention. Figure 1 As shown, method 100 may include the following steps.

[0048] Step S110, obtaining the number of rollers of the leveling machine based on the production line.

[0049] For example, the production line may be determined first, and then the number of rollers of the leveling machine may be determined based on the production line.

[0050] Step S120, determining the network structure of the budget model based on the number of rollers.

[0051] After determining the number of rollers in the leveling machine, it can be determined whether the network structure for building the budget model includes the amount of roller shifting based on the number of rollers. Figure 2 A schematic diagram of the network structure of a budget model according to an embodiment of the present invention is shown. Taking the constructed budget model as a deep neural network as an example, the deep neural network is a powerful deep learning technology that can learn and parse complex data relationships. Optionally, when constructing the budget model, the complex relationship between the bending roll force, the amount of roll shifting and the yield strength, speed and other characteristics, as well as the characteristics and process requirements of the product are fully considered. At the same time, in order to improve the generalization ability and prediction accuracy of the model, some advanced deep learning technologies such as batch normalization and dropout can be used.

[0052] Step S130, inputting the feature quantity source data into the estimation model including the network structure to train the estimation model.

[0053] It should be noted that the bending roll force and the amount of roll shifting are important process parameters of the leveling mill, which directly affect the quality of the rolled products, production efficiency and life of the equipment. The size of the bending roll force and the amount of roll shifting is affected by many factors, such as the yield strength of the rolled material, the rolling speed, the rolling temperature, the diameter of the roll and the surface roughness. Different rolled products have different material properties and process requirements, which requires the bending roll force and the amount of roll shifting budget model to be able to adapt to a variety of product characteristics and provide accurate predictions of the bending roll force and the amount of roll shifting. After obtaining the above-mentioned various data, they can be input into the budget model to train the model. Preferably, the model parameters can be optimized by the back propagation algorithm. Specifically, during the training process, a variety of optimization algorithms can be used, such as the stochastic gradient descent algorithm (SGD), the adaptive moment estimation algorithm (Adam), etc., and adjusted according to the actual situation. In addition, in order to prevent overfitting and improve the generalization ability of the model, some regularization techniques can be used, such as L1 regularization, L2 regularization, etc.

[0054] Step S140, pre-calculating the roll bending force and roll shifting amount using the trained budget model.

[0055] For example, the trained model can be used to predict the bending force of the cold rolling mill, thereby outputting the prediction result to achieve the pre-calculation of the bending force. The pre-calculation process of the roll shifting amount is similar to this and will not be repeated here. Optionally, after obtaining the prediction result, the prediction result can be compared with the actual bending force value. Figure 3 A schematic diagram showing the predicted value and actual value of the bending roll force and the roll shifting amount according to an embodiment of the present invention is shown. Figure 3 The data shown can be used to evaluate the prediction accuracy of the model and the loss function of the training process. Figure 4 FIG. 4 is a schematic diagram showing a loss function according to an embodiment of the present invention. Figure 4 The evaluation results of the loss function shown can continuously adjust the model parameters and structure so that the model can adapt to changes in different process conditions and product requirements, thereby maintaining its good prediction performance.

[0056] According to the above technical solution, the number of rollers of the leveling mill is obtained based on the production line; the network structure of the budget model is determined based on the number of rollers; the characteristic source data is input into the budget model containing the network structure to train the budget model; and the trained budget model is used to pre-calculate the bending roll force and the amount of roll shifting. Therefore, the budget model can be used to integrate the rolling process technology, product characteristics, deep learning, neural network, model training and other key fields of the cold rolling leveling mill to achieve accurate budgeting of the bending roll force and the amount of roll shifting, which will help improve the prediction accuracy of the setting values ​​of the bending roll force and the amount of roll shifting, production efficiency, product quality and equipment life, and bring significant economic benefits to the enterprise.

[0057] In addition, there is a technical problem in the existing technology that it is impossible to use the latest production and experimental data for self-learning and can only rely on manual continuous optimization and control. In this regard, it is proposed to evaluate the trained budget model and use an incremental self-learning method to update the parameters of the budget model when the evaluation result indicates low prediction accuracy.

[0058] Exemplarily, the evaluation operation may include evaluating the loss function value, prediction accuracy and other indicators of the model training process to judge the effect of the model training. If the model training effect is not good, the model parameters and structure can be adjusted continuously and retrained so that the model can adapt to the changes in different process conditions and product requirements, thereby maintaining its good prediction performance. At the same time, the performance of the model in practical applications is evaluated: the model can be applied to actual production, the prediction results can be compared with the actual rolling force value, and the prediction accuracy and stability of the model can be evaluated. For example, when the prediction result is lower than the preset threshold, it can be considered that the prediction accuracy is low, otherwise, the prediction accuracy is normal. If there are problems with the model in practical applications, data can be collected in a directional manner to enter incremental self-learning to improve the adaptability and accuracy of the model. Specifically, the model parameters can be updated by gradually introducing new data to achieve incremental self-learning. The budget model has the advantages of high precision, strong adaptability, good real-time performance and easy expansion, and can be widely used in the production control of cold rolling and leveling mills to improve production efficiency and product quality.

[0059] Optionally, when the result of the accuracy assessment indicates low accuracy, updating the parameters of the budget model using an incremental self-learning method may include: collecting relevant data of products in categories with low accuracy; and gradually inputting the relevant data into the budget model to update the parameters of the budget model.

[0060] Preferably, relevant data of products of categories with low precision can be collected. After the data is collected, the relevant data can be preprocessed to filter out noise data and abnormal data to avoid affecting the accuracy of model update and ensure the stability and reliability of the model. After that, the preprocessed relevant data can be gradually input into the budget model for incremental self-learning. Figure 5 FIG. 1 is a schematic flow chart of an incremental self-learning process for calling a flattening machine roll bending force and roll shifting amount budget model according to an embodiment of the present invention. Figure 5As shown, the entire self-learning process starts with the receipt of the steel coil output telegram as the start signal. First, the external process is executed, and the Java class is loaded to execute the executable exe file. The executable exe file starts to execute the online self-learning process. First, it is connected to the Oracle database, the deviation between the predicted value and the actual value is calculated, and the predicted value and the deviation are written into the database. Then, the historical data of the same category of products is read from the database, and the accuracy of the current model is evaluated using the above deviation and historical data to obtain the evaluation result. When the evaluation result meets the standard, the connection with the database can be terminated, thereby ending the self-learning process. On the contrary, if the evaluation result does not meet the standard, that is, when the accuracy of the current model is low, the category product corresponding to the predicted value is marked and written into the database, and then the relevant data of the category product is collected. Preferably, during the data collection process, it can be determined whether the amount of data collected is greater than or equal to 1,000, and the production line is stopped. If not, the connection with the database can also be terminated, thereby terminating the self-learning process. Otherwise, if yes, the model is incrementally self-learned, and then the above scheme is used to obtain a new prediction value, and the deviation between the new prediction value and the actual value is recalculated. It is determined again whether the model after self-learning meets the standard. If not, the above self-learning process is repeated. If it meets the standard, the data table is cleared, the connection with the database is terminated, and the self-learning process is terminated. In other words, the incremental self-learning process ends when the accuracy of the budget model meets the standard.

[0061] Therefore, the above-mentioned online learning technologies such as small batch update and learning rate decay can improve the accuracy of the model, enable the model to adapt to changes in product characteristics and process requirements, and improve the prediction accuracy and generalization ability of the model. It can continuously optimize and improve itself to adapt to more complex and diversified production needs in the future. It realizes the accurate prediction of bending roll force and roll shifting, improves production efficiency and product quality, and brings significant economic benefits to the enterprise.

[0062] Optionally, the optimized model can be embedded into the existing production line process control database and integrated into the production line process control system to achieve real-time budgeting of bending roll force and roll shifting. It is also possible to dynamically collect real-time data of products with large deviations based on the deviations evaluated by the model, dynamically optimize and adjust the model, and prepare for the next step of dynamic optimization and adjustment of the model.

[0063] According to the above technical solution, this application innovatively applies the deep neural network (DNN) model structure to the pre-calculation of the bending roll force of the cold rolling mill, describes its complex nonlinear relationship, and effectively improves the fitting ability and generalization ability of the model. The friendly user interface enables the model to be integrated into other systems such as the secondary system of the production line, realizes the monitoring and self-learning optimization mechanism of the bending roll force, improves the pre-calculation accuracy of the bending roll force of the flat mill, expands the adaptability of the model, improves the adaptive ability of the model, reduces the difficulty of maintenance for engineers, and provides pan-intelligent support for product process control in combination with the quality control needs of the market. The innovative use of speed for the direct pre-calculation of the bending roll force of the flat mill provides accurate budget data for the speed feedforward control of the bending roll force of the flat mill. The present invention can also provide strong support for the production decision-making of enterprises, help enterprises realize refined and intelligent production management, and improve the core competitiveness of enterprises.

[0064] According to the second aspect of the present invention, a device for constructing a cold rolling mill bending force and roll shifting amount estimation model is also proposed. Figure 6 FIG. 6 is a schematic block diagram of a device 600 for constructing a cold rolling mill bending force and roll shifting amount estimation model according to an embodiment of the present invention. Figure 6 As shown, the apparatus 600 may include: a roller number determination module 610 , a structure determination module 620 , a training module 630 , and a calculation module 640 .

[0065] A roller number determination module 610 is used to obtain the roller number of the leveling machine based on the production line;

[0066] A structure determination module 620, for determining a network structure of a budget model based on the number of rollers;

[0067] A training module 630, for inputting the feature quantity source data into the budget model including the network structure to train the budget model;

[0068] The calculation module 640 is used to pre-calculate the bending roll force and the roll shifting amount using the trained budget model.

[0069] According to a third aspect of the present invention, an electronic device is also provided. Figure 7 FIG. 8 is a schematic block diagram of an electronic device 700 according to an embodiment of the present invention. Figure 7 As shown, the electronic device 700 may include a processor 710 and a memory 720. The memory 720 stores computer program instructions, which are used by the processor 710 to execute the method for constructing the cold rolling mill bending force and roll shifting amount budget model as described above.

[0070] According to a fourth aspect of the present invention, a storage medium is also provided, on which program instructions are stored, and the program instructions are used to execute the method for constructing the cold rolling mill bending roll force and roll shifting amount budget model as described above when running. The storage medium may include, for example, a storage component of a tablet computer, a hard disk of a computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

[0071] A person of ordinary skill in the art can understand the specific details and beneficial effects of the construction device, electronic device and storage medium of the cold rolling mill bending roll force and roll shifting amount budget model by reading the above description of the method for constructing the cold rolling mill bending roll force and roll shifting amount budget model, which will not be elaborated here for the sake of brevity.

[0072] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing the following method steps:

[0073] Get the number of rollers of the skin-pass mill based on the production line;

[0074] Determine the network structure of the budget model based on the number of rollers;

[0075] Inputting the feature quantity source data into the budget model including the network structure to train the budget model;

[0076] The trained budget model is used to pre-calculate the bending roll force and roll shifting amount.

[0077] Optionally, the method further comprises:

[0078] The trained budget model is evaluated to update parameters of the budget model using an incremental self-learning method when the evaluation result indicates low prediction accuracy.

[0079] Optionally, when the result of the accuracy evaluation indicates low accuracy, an incremental self-learning method is used to update parameters of the budget model, including:

[0080] Collect relevant data of products with low accuracy;

[0081] Gradually input relevant data into the budget model to update the parameters of the budget model

[0082] Optionally, before gradually inputting the relevant data into the budget model, the method further comprises:

[0083] Preprocess relevant data.

[0084] Optionally, the method further comprises:

[0085] Adjust the network structure of the budget model and / or increase training data based on the evaluation results.

[0086] Optionally, inputting the feature quantity source data into a budget model including a network structure to train the budget model includes:

[0087] The budget model parameters are optimized through the back-propagation algorithm.

[0088] Optionally, training the budget model further includes:

[0089] Regularization techniques are used during the training of the budget model.

[0090] In the several embodiments provided in the present application, it should be understood that the disclosed devices and / or equipment can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as 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 devices or units, which can be electrical, mechanical or other forms.

[0091] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0093] 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 application is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0094] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a cold rolling mill roll bending force and roll shifting amount estimation model, characterized in that: include: Obtaining the number of rollers of the skin-pass mill based on the production line; determining a network structure of the budget model based on the number of rollers; Inputting feature quantity source data into the budget model including the network structure to train the budget model; The trained budget model is used to pre-calculate the bending roll force and the roll shifting amount.

2. The method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to claim 1, characterized in that: The method further comprises: The trained budget model is evaluated, and when the evaluation result indicates that the prediction accuracy is low, the parameters of the budget model are updated using an incremental self-learning method.

3. The method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to claim 2, characterized in that: When the evaluation result indicates that the prediction accuracy is low, an incremental self-learning method is used to update the parameters of the budget model, including: Collect relevant data of the low-precision products; The relevant data is gradually input into the budget model to update the parameters of the budget model.

4. The method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to claim 3, characterized in that: Before gradually inputting the relevant data into the budget model, the method further includes: The relevant data are preprocessed.

5. The method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to claim 2, characterized in that: The method further comprises: The network structure of the budget model is adjusted and / or training data is added according to the evaluation results.

6. The method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to claim 1, characterized in that: The step of inputting the feature quantity source data into the budget model including the network structure to train the budget model includes: The budget model parameters are optimized by a back-propagation algorithm.

7. The method for constructing a cold rolling skin-pass mill bending force and roll shifting amount estimation model according to claim 6, characterized in that: The training of the budget model further comprises: Regularization techniques are employed during the training of the budget model.

8. A device for constructing a cold rolling mill roll bending force and roll shifting amount estimation model, characterized in that: include: A roller number determination module, used for obtaining the roller number of the leveling machine based on the production line; A structure determination module, used for determining the network structure of the budget model based on the number of rollers; A training module, used for inputting feature quantity source data into the budget model including the network structure to train the budget model; A calculation module is used to pre-calculate the bending roll force and the roll shifting amount by using the trained budget model.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions are used by the processor to execute the method for constructing a cold rolling mill bending roll force and roll shifting amount budget model as claimed in any one of claims 1 to 7 when the processor is running.

10. A storage medium having program instructions stored thereon, wherein the program instructions are used to execute, when running, the method for constructing a cold rolling mill bending roll force and roll shifting amount estimation model as claimed in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Setting method and device for roll bending force of six-roller cold-rolling mill

    CN106216409A

  • Cold-rolled sheet shape control parameter setting method based on error back propagation algorithm

    CN114192587A

  • Cold continuous rolling roll bending force online prediction method based on data driving

    CN116174497A

  • Method for forecasting and controlling width narrowing amount of strip steel in whole cold rolling process

    CN116984380A

  • Prediction method of crown of steel plates and strips based on data driving and mechanism model fusion

    US20240184956A1