Rolling method, system and equipment in full-closed mode and storage medium

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to a rolling method, system and equipment in a full-closed mode and a storage medium, and the method comprises the steps that plate thickness and temperature data are collected in real time through a thickness measuring device integrated with a temperature sensor; predicting a future thickness change trend by using a temperature-sensitive deep learning model and combining historical working condition parameters; calculating a dynamic compensation gap sequence based on the prediction result, the target thickness and the compression roller characteristics, and driving a compression roller control system to adjust the gap; and feedback data after rolling are synchronously collected, model parameters are updated in a self-adaptive mode, and closed-loop control is formed. According to the full-closed mode rolling method, high-precision rolling is achieved through closed-loop control, and the quality and efficiency are improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent manufacturing technology, and in particular to a fully closed mode rolling method, system, equipment and storage medium. Background Art

[0002] In the field of sheet metal rolling, particularly in the production of high-precision alloy sheets for aerospace applications, semiconductor-grade foils, and dissimilar metal composites, thickness accuracy is a core quality indicator. During the rolling process, sheet thickness is influenced by a variety of factors, including temperature variations (which affect the thermal expansion coefficient and yield strength of the material), the mechanical properties of the rollers (such as roll gap deviation caused by thermal expansion), and material variations. Traditional processing methods face key technical challenges.

[0003] Existing technologies primarily adjust the roll gap through manual experience or fixed models based on physical formulas, which can meet the production needs of some conventional sheet metal products. However, faced with complex operating conditions such as sudden temperature changes during high-speed rolling, the uncertainty of the mechanical properties of new materials, and the interlaminar deformation differences of dissimilar metal composite sheets, traditional methods have significant limitations. First, they lack effective perception of the real-time coupled effects of multiple variables such as temperature, material, and equipment status, making it difficult to dynamically adapt to real-time changes in the material's thermal expansion coefficient and yield strength. Second, they rely on empirical formulas or linear models for thickness prediction, which are insufficient to adapt to nonlinear and time-varying rolling processes, making it difficult to ensure accurate predictions for new materials or complex operating conditions. Third, roll gap adjustment strategies lag, often relying on post-compensation or fixed timing control. They cannot proactively address thickness trends, which can easily lead to batch-to-batch deviations. Fourth, they lack a closed-loop calibration mechanism, lacking adaptability to the long-term effects of external factors such as equipment thermal deformation and environmental parameters (such as humidity and air pressure), making it difficult to achieve high-precision and stable control.

[0004] With the continuous improvement of sheet metal quality requirements in high-end manufacturing and the emergence of new materials and operating conditions, the shortcomings of traditional open-loop control models in various implementation aspects have become increasingly prominent. The key technical challenge facing the sheet metal rolling industry is how to build a fully closed-loop control system driven by real-time data to accurately predict and proactively compensate for temperature-sensitive thickness variations. Summary of the Invention

[0005] The purpose of this application is to provide a fully closed mode rolling method, system, equipment and storage medium, aiming to solve the problem of thickness accuracy control caused by the lack of closed-loop control capability during plate rolling.

[0006] The first object of this application is to provide a fully closed mode rolling method, comprising: The real-time thickness data of the plate to be rolled and the real-time temperature data of the corresponding position are obtained in real time through the thickness measuring device with integrated temperature sensor; Preset a temperature-sensitive deep learning model, which takes the real-time temperature data, the real-time thickness data and historical rolling condition parameters as inputs, and outputs a prediction matrix of the thickness change trend of the sheet under different temperature evolution paths within a preset future time; Based on the prediction matrix of the thickness change trend of the sheet, combined with the target thickness of the sheet and the current mechanical characteristics parameters of the pressure roller, calculate a dynamic compensation pressure roller gap sequence, and the dynamic compensation pressure roller gap sequence includes the gap adjustment amount and adjustment timing of several future control cycles; Send the dynamic compensation pressure roller gap sequence to the pressure roller control system, and the pressure roller control system dynamically adjusts the pressure roller gap, synchronously collects the thickness feedback data and the corresponding temperature feedback data of the sheet after the pressure roller is adjusted and the rolling is completed, and uploads the thickness feedback data and the temperature feedback data.

[0007] By adopting the above technical solution, the real-time synchronous acquisition of the sheet thickness and temperature data is realized through the thickness measuring device integrated with the temperature sensor, ensuring the dynamic perception ability of the key parameters in the rolling process and providing real-time data support for precise control; the introduction of the temperature-sensitive deep learning model breaks through the limitations of traditional empirical formulas or linear models, can effectively capture the non-linear mapping relationship between temperature evolution and sheet thickness change, and significantly improves the prediction ability of the thickness change trend under complex working conditions (such as sudden temperature change, new material rolling); the dynamic compensation pressure roller gap sequence generated based on the prediction matrix can be adjusted prospectively according to the thickness change trend in multiple future cycles. Compared with the traditional lagging after-the-fact compensation mechanism, it realizes the dynamic optimization and early adaptation of the pressure roller gap, effectively suppressing the thickness deviation caused by factors such as temperature fluctuation and material property change; by incorporating the thickness and temperature feedback data after rolling into the closed-loop control process, a complete closed-loop of "data acquisition - model prediction - gap adjustment - feedback calibration" is formed, enabling the system to continuously optimize the control parameters according to the actual rolling effect, enhancing the self-adaptability to long-term influences such as equipment state changes and environmental parameter fluctuations, and fundamentally improving the intelligent level and thickness control accuracy of the metal sheet rolling process, providing an innovative solution for high-precision and high-stability modern rolling production.

[0008] In a possible implementation manner of the present application, the method includes: Integrate a spectral detector in the thickness measuring device integrated with the temperature sensor, collect the surface spectral data of the sheet in real time, and online identify the material type and mechanical property parameters of the currently rolled sheet through a preset material identification model; According to the identified material type and mechanical property parameters, retrieve the corresponding material-specific prediction sub-model; If the sheet material is a new material and there is insufficient historical data, transfer learning technology is adopted. Based on the exclusive prediction sub-model for similar materials, the parameters of the temperature-sensitive deep learning model are adjusted in combination with the mechanical property parameters; When calculating the dynamic compensation roll gap sequence, a safety margin compensation term is added for new materials or materials with insufficient data.

[0009] By adopting the above technical solutions, by integrating a spectral detector and a material identification model in the thickness measurement device integrated with a temperature sensor, real-time online identification of the sheet material type and mechanical property parameters is achieved, avoiding the hysteresis and subjectivity brought by manual intervention, and providing a data basis for accurately calling the exclusive prediction sub-model for materials; the setting of the exclusive prediction sub-model for materials can perform customized prediction according to the characteristics of different materials, significantly improving the prediction accuracy of the thickness change trend of specific materials during the rolling process compared with the general model; for new materials or insufficient historical data, the application of transfer learning technology can quickly build an adapted model based on the historical knowledge of similar materials, solving the problem of prediction failure caused by data scarcity in traditional methods and greatly shortening the process debugging cycle of new materials; the addition of the safety margin compensation term enhances the fault tolerance ability of the system to material uncertainty, and the thickness over-tolerance risk is avoided in advance through the compensation mechanism when the model prediction reliability is insufficient, thus improving the adaptability and stability of the rolling process to new materials and complex materials, and providing a universal solution for the high-precision rolling of diverse materials.

[0010] In a possible implementation manner of this application, the method includes: Real-time obtain the humidity, air pressure and equipment vibration data of the rolling environment through the environmental sensors set at the rolling site; Analyze the influence laws of the humidity, air pressure and equipment vibration data on the temperature transfer and thickness change during the sheet rolling process, and construct an environmental correction model; Input the humidity, air pressure and equipment vibration data, the real-time temperature data, the real-time thickness data, the historical rolling condition parameters, the material type and the mechanical property parameters into the temperature-sensitive deep learning model or the exclusive prediction sub-model for materials; Correct the prediction matrix of the sheet thickness change trend.

[0011] By adopting the above technical solutions, environmental parameters such as humidity, air pressure, and equipment vibration are collected in real time through environmental sensors, an environmental correction model is constructed and incorporated into the temperature-sensitive deep learning model, realizing dynamic quantitative analysis and precise compensation for the influence of environmental factors during the rolling process: the acquisition of real-time environmental data enables the system to perceive the influence of humidity on the heat dissipation efficiency of the sheet surface, the effect of air pressure on the atomization effect of the cooling medium, and the interference of equipment vibration on the stability of the pressure roller, avoiding prediction errors caused by the neglect or simplified treatment of environmental factors in traditional methods; the introduction of the environmental correction model quantifies the correlation law between environmental parameters and temperature transfer and thickness change, filling the modeling gap of the coupled action of multiple physical fields in a complex environment; inputting environmental data, material parameters, and real-time working condition data into the intelligent model together enables the prediction matrix to dynamically adapt to environmental changes (such as the decrease in the cooling rate of the sheet in a high-humidity environment and the small fluctuations in the pressure roller gap under high-vibration working conditions), effectively correcting the thickness prediction error caused by environmental fluctuations, improving the generalization ability and prediction accuracy of the model under different environmental conditions, thereby enhancing the adaptability of the rolling system to non-ideal working conditions such as changes in workshop temperature and humidity and long-term vibration deviation of equipment during operation, and providing environmental robustness guarantee for the stable realization of high-precision thickness control.

[0012] In a possible implementation manner of the present application, the method includes: When it is monitored that the temperature change rate of the sheet exceeds the threshold, based on the material type, the mechanical property parameters, the target thickness, and the humidity, air pressure, and equipment vibration data of the current sheet, calculate the optimal stable temperature range through the temperature-sensitive deep learning model; Adjust the temperature of the sheet to the optimal stable temperature range through the heating device or cooling device integrated on the rolling line, and the adjustment power of the heating device or cooling device is dynamically matched according to the real-time heat radiation data of the sheet; Synchronously input the adjusted temperature data, the material type, the mechanical property parameters, the humidity, air pressure, and equipment vibration data after the sheet temperature is adjusted stably into the temperature-sensitive deep learning model to generate a low-fluctuation exclusive prediction matrix; Calculate the dynamic compensation pressure roller gap sequence based on the low-fluctuation exclusive prediction matrix.

[0013] By adopting the above technical solution, after detecting that the temperature change rate of the sheet exceeds the limit, the optimal stable temperature range is calculated based on the material properties, mechanical parameters and environmental data, and combined with the dynamic power adjustment of the heating / cooling device, active intervention and precise control of the drastic temperature fluctuation are realized: the temperature regulation driven by real-time thermal radiation data ensures that the temperature of the sheet quickly stabilizes in the range where the material deformation resistance is low and the thermal expansion coefficient is stable, avoiding the interference of the sudden change of yield strength caused by the sudden temperature change and the uneven thermal expansion on the thickness control; the stabilized temperature data, environmental and material parameters are synchronously input into the model to generate a low-fluctuation exclusive prediction matrix, effectively filtering out the noise introduced by the high-frequency temperature fluctuation, making the prediction matrix more accurately reflect the deformation law of the material in the stable thermal state; the dynamic compensation roll gap sequence calculated based on this matrix can accurately match the material flow characteristics after the temperature is stabilized, significantly reducing the thickness prediction deviation and compensation lag caused by the unstable temperature, improving the thickness control accuracy of heat-sensitive materials such as superalloys and titanium alloys during high-speed rolling, providing a closed-loop control strategy for solving the problem of uneven sheet deformation under the condition of drastic temperature change, and enhancing the adaptability and stability of the rolling system to extreme temperature changes.

[0014] In a possible implementation manner of the present application, the method includes: When the rolled sheet is a dissimilar metal composite sheet, an ultrasonic flaw detector and a metallographic microscope probe are added to the thickness measuring device of the integrated temperature sensor to obtain the bonding condition data of the interface of the dissimilar metal composite sheet in real time, and the bonding condition data includes bonding strength information and interface defect information; Spectral analysis is performed on each layer of metal of the dissimilar metal composite sheet, and the material type and mechanical property parameters of each layer of metal are accurately identified through a preset material identification model; At the same time, a comprehensive mechanical property model of the dissimilar metal composite sheet is established, considering the interaction between each layer of metal to more accurately describe the overall mechanical behavior of the dissimilar metal composite sheet; According to the material type, the mechanical property parameters and the bonding condition data of each layer of metal, a customized temperature-sensitive deep learning sub-model is developed. The temperature-sensitive deep learning sub-model focuses on the heat transfer and deformation coordination relationship between different metal layers during the training process to improve the prediction accuracy of the thickness change trend of the dissimilar metal composite sheet; When calculating the dynamic compensation roll gap sequence, the target thickness of the sheet and the mechanical characteristic parameters of the roll are considered, and combined with the bonding condition data and the comprehensive mechanical property model of the dissimilar metal composite sheet, an additional gap compensation amount is added for the weak bonding area of the interface to ensure uniform deformation and good bonding of each layer of metal of the dissimilar metal composite sheet during the rolling process.

[0015] By adopting the above technical solutions, aiming at the problem of interlayer coordination of dissimilar metal composite plates, through the addition of ultrasonic flaw detectors and metallographic microscope probes, the real-time and accurate monitoring of the interface bonding strength and defects is achieved, avoiding the risk of interlayer peeling caused by unknown interface states in traditional methods; the combination of spectral analysis and material identification technology ensures the accurate acquisition of the material properties and mechanical parameters of each layer of metal, providing data support for constructing a comprehensive mechanical property model considering interlayer interaction; the customized temperature-sensitive deep learning sub-model focuses on the interlayer heat transfer and deformation coordination relationship, breaking through the simplification limitation of the traditional model in describing the overall mechanical behavior of composite plates, and significantly improving the prediction accuracy of the thickness change trend; the design of additional gap compensation amount for weak interface bonding areas can dynamically balance the deformation resistance differences of each layer of metal (such as interlayer stress caused by different thermal expansion coefficients and yield strengths), promoting the uniform deformation of each layer of metal during rolling, effectively reducing defects such as interface cracks and peeling caused by interlayer stress concentration, ensuring the interface bonding quality and overall dimensional accuracy of composite plates, and providing a targeted solution for the high-precision manufacturing of dissimilar metal composite plates.

[0016] In a possible implementation manner of the present application, the method includes: Real-time obtain the temperature change data of each layer of metal through a plurality of high-precision temperature sensors arranged at the rolling site; Calculate the thermal expansion amounts at different positions according to the thermal expansion coefficients of each layer of metal, and generate a thermal expansion distribution map; For the thermal expansion amount differences of each layer of metal in the thermal expansion distribution map, determine the metal layers that need to be temperature-controlled and their control directions; Independently control the temperature of each layer of metal through the heating device or cooling device integrated on the rolling line; Real-time monitor the changes in the thermal expansion amounts of each metal layer, and adjust the heating or cooling rate according to the changes in the thermal expansion amounts; Obtain the thermal expansion monitoring data of each metal layer; Obtain the temperature control parameters of the heating device or cooling device; Input the thermal expansion monitoring data and the temperature control parameters into the temperature-sensitive deep learning sub-model, and update the prediction matrix of the thickness change trend of the plate in real time to more accurately reflect the influence of thermal expansion differences on the thickness and interface bonding of the dissimilar metal composite plate.

[0017] By adopting the above technical solutions, aiming at the deformation coordination problem caused by the interlayer thermal expansion difference of dissimilar metal composite plates, the temperature changes of each layer of metal are captured in real time by high-precision temperature sensors, and combined with the calculation of the coefficient of thermal expansion and the visualization of the distribution, the accurate quantitative analysis of the interlayer thermal deformation difference is realized; based on the independent temperature control strategy of the thermal expansion distribution map, the temperature of each metal layer can be adjusted specifically (such as cooling the metal layer with a large thermal expansion and heating and compensating the layer with a small expansion), dynamically balancing the interlayer thermal expansion amount, and effectively reducing the interfacial shear stress caused by the expansion difference; the change of the thermal expansion amount is monitored in real time and fed back to the temperature-sensitive deep learning sub-model, so that the prediction matrix can reflect the change of the mechanical behavior after the interlayer thermal state is adjusted in real time, avoiding the prediction deviation caused by ignoring the dynamic process of interlayer heat transfer in the traditional method; this mechanism promotes the deformation uniformity of each metal layer during the rolling process, reduces defects such as cracks and peeling caused by thermal stress concentration in the interface bonding area, improves the overall thickness accuracy and interface bonding quality of the composite plate, and provides the key technical support for the interlayer thermal coordination control in the high-precision and high-reliability manufacturing of dissimilar metal composite plates.

[0018] In a possible implementation manner of the present application, the method includes: Embed distributed fiber Bragg grating sensors inside the roll body and bearing housing of the pressure roll, collect the temperature distribution data of the circumferential and axial directions of the pressure roll in real time, and obtain the surface temperature gradient of the pressure roll and the temperature rise rate of the bearing housing; Combined with the coefficient of thermal expansion of the pressure roll material, construct a thermal deformation prediction model of the pressure roll, and calculate the gap deviation caused by the radial expansion of the pressure roll due to temperature change based on the heat conduction equation and the expansion formula; Input the temperature distribution data, the gap deviation, the real-time thickness data and the real-time temperature data into the temperature-sensitive deep learning model together to generate a thermal compensation dynamic gap sequence; Collect the thickness feedback data of the rolled plate through the thickness measuring device. If the thickness deviation exceeds the preset range within a continuous preset period and the deviation direction is consistent with the predicted direction of the thermal deformation of the pressure roll, extract the actual temperature of the pressure roll, the actual gap between the pressure roll and the plate, and the actual thickness of the plate at the current moment, and update the weight parameter of the thermal deformation of the pressure roll in the temperature-sensitive deep learning model to realize the adaptive optimization of the compensation strategy.

[0019] By adopting the above technical solutions, precise monitoring and adaptive compensation control of the thermal deformation of the pressure roller are achieved. The distributed fiber Bragg grating sensors collect the circumferential and axial temperature distribution data of the pressure roller in real time, and can accurately capture the surface temperature gradient of the pressure roller and the temperature rise rate of the bearing housing, providing a basis for constructing an accurate prediction model of the thermal deformation of the pressure roller. The model constructed based on the heat conduction equation and the expansion formula can calculate the gap deviation caused by the radial expansion of the pressure roller, enabling the system to predict in advance the influence of thermal deformation on the rolling gap. By inputting data such as temperature distribution and gap deviation into the temperature-sensitive deep learning model to generate a thermal compensation dynamic gap sequence, the gap of the pressure roller can be adjusted in time to compensate for thermal deformation, improving the control accuracy of the rolling thickness. When the thickness deviation of the rolled sheet exceeds the preset range within a continuous preset period after rolling and is consistent with the predicted direction of the thermal deformation of the pressure roller, relevant actual data is extracted to update the weight parameters of the thermal deformation of the pressure roller in the model, enabling the adaptive optimization of the compensation strategy, allowing the system to maintain a good thickness control effect under different working conditions and during long-term operation, enhancing the stability and reliability of the rolling process, reducing the reject rate caused by the thermal deformation of the pressure roller, and improving the product quality and production efficiency.

[0020] The second object of this application is to provide a rolling system in a fully closed mode, and this system includes: Thickness and temperature data acquisition module: Real-time thickness data of the sheet to be rolled and real-time temperature data at the corresponding position are acquired in real time through a thickness measuring device integrated with temperature sensors; Sheet thickness change trend prediction matrix output module: A temperature-sensitive deep learning model is preset, and with the real-time temperature data, the real-time thickness data, and historical rolling condition parameters as inputs, a prediction matrix of the sheet thickness change trend under different temperature evolution paths within a future preset time is output; Dynamic compensation pressure roller gap sequence calculation module: Based on the prediction matrix of the sheet thickness change trend, combined with the target thickness of the sheet and the current mechanical characteristic parameters of the pressure roller, a dynamic compensation pressure roller gap sequence is calculated, and the dynamic compensation pressure roller gap sequence includes the gap adjustment amounts and adjustment time sequences for several future control cycles; Pressure roller adjustment and feedback data uploading module: The dynamic compensation pressure roller gap sequence is sent to the pressure roller control system, and the pressure roller control system dynamically adjusts the gap of the pressure roller, synchronously collects the thickness feedback data and the corresponding temperature feedback data of the sheet after the pressure roller is adjusted and the rolling is completed, and uploads the thickness feedback data and the temperature feedback data.

[0021] By adopting the above technical solutions, the thickness measurement device integrated with a temperature sensor realizes the real-time synchronous acquisition of sheet thickness and temperature data, ensuring the dynamic perception ability of key parameters in the rolling process and providing real-time data support for precise control; the introduction of a temperature-sensitive deep learning model breaks through the limitations of traditional empirical formulas or linear models, and can effectively capture the non-linear mapping relationship between temperature evolution and sheet thickness changes, significantly improving the prediction ability of thickness change trends under complex working conditions (such as sudden temperature changes and rolling of new materials); the dynamic compensation roll gap sequence generated based on the prediction matrix can be adjusted prospectively according to the thickness change trends in future multiple cycles. Compared with the traditional lagging ex-post compensation mechanism, it realizes the dynamic optimization and early adaptation of the roll gap, effectively suppressing thickness deviations caused by factors such as temperature fluctuations and material property changes; by incorporating the thickness and temperature feedback data after rolling into the closed-loop control process, a complete closed-loop of "data acquisition - model prediction - gap adjustment - feedback calibration" is formed, enabling the system to continuously optimize control parameters according to the actual rolling effect, enhancing the self-adaptability to long-term impacts such as changes in equipment status and fluctuations in environmental parameters, and fundamentally improving the intelligent level and thickness control accuracy of the metal sheet rolling process, providing an innovative solution for modern rolling production with high precision and high stability.

[0022] The third objective of this application is to provide a rolling device in a fully closed mode, which includes: A memory and a processor, and a computer program capable of being loaded and executed by the processor for the above-mentioned fully closed mode rolling method is stored on the memory.

[0023] The fourth objective of this application is to provide a storage medium.

[0024] The above fourth objective of this application is achieved through the following technical solutions: A storage medium, in which a computer program capable of being loaded and executed by the processor for the above-mentioned fully closed mode rolling method is stored.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. The real-time synchronous acquisition of sheet thickness and temperature data is achieved through a thickness measurement device integrated with a temperature sensor, ensuring the dynamic perception ability of key parameters in the rolling process and providing real-time data support for precise control. The introduction of a temperature-sensitive deep learning model breaks through the limitations of traditional empirical formulas or linear models, and can effectively capture the non-linear mapping relationship between temperature evolution and sheet thickness change, significantly improving the prediction ability of thickness change trends under complex working conditions (such as sudden temperature changes, rolling of new materials). The dynamic compensation roll gap sequence generated based on the prediction matrix can be adjusted prospectively according to the thickness change trends in future multiple cycles. Compared with the traditional lagging ex-post compensation mechanism, it realizes the dynamic optimization and early adaptation of the roll gap, effectively suppressing thickness deviations caused by factors such as temperature fluctuations and changes in material properties. By incorporating the thickness and temperature feedback data after rolling into the closed-loop control process, a complete closed-loop of "data acquisition - model prediction - gap adjustment - feedback calibration" is formed, enabling the system to continuously optimize control parameters according to the actual rolling effect, enhancing the self-adaptability to long-term influences such as changes in equipment status and fluctuations in environmental parameters, and fundamentally improving the intelligent level and thickness control accuracy of the metal sheet rolling process, providing an innovative solution for high-precision and high-stability modern rolling production.

[0026] 2. Aiming at the problem of interlayer coordination of dissimilar metal composite sheets, by adding an ultrasonic flaw detector and a metallographic microscope probe, the real-time and accurate monitoring of the interface bonding strength and defects is achieved, avoiding the risk of interlayer peeling caused by unknown interface states in traditional methods. The combination of spectral analysis and material identification technology ensures the accurate acquisition of the material properties and mechanical parameters of each layer of metal, providing data support for constructing a comprehensive mechanical property model considering interlayer interactions. The customized temperature-sensitive deep learning sub-model focuses on the interlayer heat transfer and deformation coordination relationship, breaking through the simplification limitations of traditional models in describing the overall mechanical behavior of composite sheets, and significantly improving the prediction accuracy of thickness change trends. The design of additional gap compensation amount for weak interface bonding areas can dynamically balance the differences in deformation resistance of each layer of metal (such as interlayer stress caused by different thermal expansion coefficients and yield strengths), promoting the uniform deformation of each layer of metal during rolling, effectively reducing defects such as interface cracks and peeling caused by interlayer stress concentration, and ensuring the interface bonding quality and overall dimensional accuracy of the composite sheet, providing a targeted solution for the high-precision manufacturing of dissimilar metal composite sheets. Description of the Drawings

[0027] Figure 1 is a schematic flow chart of a full-closed mode rolling method provided by an embodiment of the present application; Figure 2 is a schematic virtual structure diagram of a full-closed mode rolling system provided by an embodiment of the present application. Detailed Embodiments

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0029] In addition, the term "and / or" in this document merely describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0030] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings of the specification.

[0031] The embodiments of the present application provide a rolling method in a fully closed mode. Referring to Figure 1 , the main process of the method is described as follows: S1: Real-time thickness data of the sheet to be rolled and real-time temperature data at the corresponding position are obtained in real time through a thickness measuring device integrated with a temperature sensor; Among them, the thickness measuring device integrated with a temperature sensor is not a simple independent device, but realizes data synchronization through hardware integration and algorithm calibration. For example, in a certain cold rolling mill, the thickness gauge measures the thickness by the laser triangulation method, and the built-in infrared array sensor synchronously collects the temperature at the corresponding position. The two are strictly aligned through time stamps and spatial coordinates to ensure that each thickness data point matches the real-time temperature value. This three-dimensional data matrix of "position-temperature-thickness" provides input with physical meaning for subsequent model analysis and avoids the data misalignment problem caused by traditional separate acquisition.

[0032] S2: A temperature-sensitive deep learning model is preset, and the real-time temperature data, the real-time thickness data, and historical rolling condition parameters are used as inputs to output a prediction matrix of the thickness change trend of the sheet under different temperature evolution paths within a preset future time; Among them, the core advantage of the temperature-sensitive deep learning model lies in dealing with the non-linear and time-varying rolling process. When the sheet temperature rapidly rises from 800°C to 850°C, the traditional linear model assumes a constant coefficient of thermal expansion and predicts a thickness increase of 0.02 mm. However, through training with historical data, this model finds that the decrease in the yield strength of the material in the high-temperature zone will additionally cause a thickness reduction of 0.01 mm, and finally outputs a more accurate prediction matrix. This capture of the complex relationship of "temperature-material properties-deformation" depends on the model's learning of a large amount of historical working conditions (such as different combinations of rolling speeds and reduction ratios), and realizes full-scenario coverage of future multi-path temperature evolution (such as uniform heating and stepwise cooling).

[0033] S3: Based on the predicted matrix of the sheet thickness change trend, combined with the target thickness of the sheet and the current mechanical characteristic parameters of the pressure roller, calculate a dynamic compensation pressure roller gap sequence, and the dynamic compensation pressure roller gap sequence includes the gap adjustment amounts and adjustment time sequences for several future control cycles; Among them, the generation of the dynamic compensation pressure roller gap sequence does not solely rely on the prediction result, but is an intelligent decision that integrates the physical characteristics of the equipment. For example, when the model predicts that the edge of the sheet will thicken by 0.03 mm, the system needs to simultaneously consider the mechanical stiffness of the pressure roller. If the stiffness in the middle of the pressure roller is higher than that at both ends, the actual rolling-down effect of the same gap adjustment amount in the edge area is more significant. Therefore, the compensation amount in the edge area of the gap sequence will be automatically adjusted (such as increasing by 0.005 mm) to ensure the consistency between the theoretical calculation and the actual response of the equipment. This coupled calculation of "model prediction + equipment characteristics" avoids the problem of the disconnection between the ideal model and the actual equipment in the traditional method.

[0034] S4: Send the dynamic compensation pressure roller gap sequence to the pressure roller control system, and the pressure roller control system dynamically adjusts the pressure roller gap, synchronously collects the thickness feedback data and the corresponding temperature feedback data of the sheet after the pressure roller adjustment and the rolling is completed, and uploads the thickness feedback data and the temperature feedback data.

[0035] Among them, after the dynamic compensation roll gap sequence is generated, it will be quickly and stably sent to the roll control system through the industrial Ethernet. The core of the roll control system is the electro-hydraulic servo mechanism, which has a high response speed and precise position control ability. Taking a common four-high rolling mill as an example, when receiving the gap adjustment instruction, the electro-hydraulic servo valve will quickly adjust the flow rate and pressure of the hydraulic oil to drive the screw-down or hydraulic cylinder to accurately move the position of the roll, ensuring that the gap adjustment is in place. After the roll completes the gap adjustment and rolls the sheet, the thickness measuring device integrated with a temperature sensor will come into play again. It will immediately measure the rolled sheet, synchronously collecting the thickness feedback data and the corresponding temperature feedback data. The acquisition accuracy of the thickness feedback data is crucial. For example, using laser interferometry thickness measurement technology, the measurement accuracy can be controlled within ±0.005 mm. At the same time, the temperature sensor will synchronously record the temperature of the corresponding position of the sheet, such as the temperature in the middle of the sheet is 55°C. These feedback data will be uploaded to the database of the control system in real time, providing a basis for subsequent analysis and calibration.

[0036] The system will perform real-time analysis on the uploaded feedback data. If there are thickness deviations in consecutive multiple cycles (such as 5 consecutive cycles), and the deviation direction is consistent with the prediction direction of the previous temperature-sensitive deep learning model, this indicates that there may be certain errors in the model prediction and calibration is required. At this time, the system will extract the "roll temperature - actual gap - sheet thickness" data pair at the current moment. For example, the temperature in the middle of the roll is 60°C, the actual gap is 2.55 mm, and the measured thickness in the middle of the sheet is 2.52 mm.

[0037] Then, the system will use an online learning algorithm (such as the stochastic gradient descent algorithm) to update the weight parameters of the temperature-sensitive deep learning model. The online learning algorithm can adjust the model in real time according to the newly collected data, enabling the model to quickly adapt to various changes in the rolling process. By continuously updating the model parameters, the subsequent prediction accuracy will be optimized, thus realizing the complete closed-loop control of prediction - adjustment - feedback - calibration. This closed-loop control mechanism can effectively improve the stability of the rolling process and the product quality, reduce the thickness deviation caused by various factors (such as material property fluctuations, environmental changes, equipment wear, etc.), and ensure that the thickness of the sheet is always controlled within the target range.

[0038] In another embodiment, for the fully closed mode rolling method, the technological innovation can be further enhanced through multi-dimensional data fusion and intelligent strategy upgrade, which is applicable to harsh scenarios such as high-precision electronic foils and aerospace ultra-thin sheets: First, introduce a high-density sensor array in the data acquisition link. In addition to real-time thickness and temperature detection, synchronously collect the stress distribution and micro-displacement data on the surface of the sheet to form a multi-modal data set of "thickness - temperature - stress - displacement". For example, in the rolling of ultra-thin copper foils, detection points are arranged at micron-level intervals along the width direction of the sheet to accurately capture the material softening areas caused by local frictional overheating, providing richer physical feature inputs for subsequent predictions and solving the problem of microscopic deformation differences that cannot be identified by traditional single-point detection.

[0039] Secondly, upgrade the temperature-sensitive model to an adaptive learning framework. By integrating multi-source data such as equipment vibration signals and hydraulic system pressure fluctuations, construct a closed-loop optimization mechanism of "prediction - decision - verification". The model can not only output the thickness change trend, but also dynamically calculate the stress safety thresholds of each region. When it is predicted that the stress in the edge region is close to the damage critical value, automatically adjust the local roll gap to avoid sheet tearing while ensuring thickness accuracy, significantly improving the control robustness under complex working conditions.

[0040] In the dynamic compensation strategy, introduce digital twin pre-verification technology, and based on the physical model of the roll and the sheet, real-time simulate the rolling effects of different gap adjustment schemes. For example, when the system generates a gap adjustment sequence, first verify through the virtual model whether this scheme will cause edge thickness out-of-tolerance or stress concentration. If there are risks, automatically optimize the adjustment amplitude and timing, transforming the physical trial-and-error of traditional trial rolling into pre-verification in the digital space to ensure the safety and accuracy of the control scheme, especially suitable for zero-risk commissioning of new materials during the first rolling.

[0041] Finally, realize the local real-time processing of feedback data through edge computing technology, sink computing tasks such as thickness deviation analysis and equipment status monitoring to the mill site nodes, and shorten the data transmission and response delay. For example, when an abnormal high-frequency vibration signal is detected, the edge node can quickly identify it as an early feature of roll bearing wear, immediately trigger a maintenance warning and adaptively adjust the control parameters, improving the system's response speed to disturbances such as equipment aging and environmental fluctuations to the sub-millisecond level, and further enhancing the accuracy stability during long-term operation.

[0042] Through the expansion of detection dimensions, intelligent upgrade of the model, virtual-real fusion verification, and edge computing enhancement, form an all-round intelligent improvement from data acquisition to control execution, effectively solving problems such as insufficient microscopic deformation monitoring, poor adaptability to extreme working conditions, and lagging equipment response in high-end sheet rolling.

[0043] Specifically, in some possible embodiments, the method includes: Integrate a spectral detector in the thickness measurement device of the integrated temperature sensor to collect spectral data on the surface of the sheet in real time, and online identify the material type and mechanical property parameters of the currently rolled sheet through a preset material identification model; According to the identified material type and mechanical property parameters, retrieve the corresponding material-specific prediction sub-model; If the sheet is a new material and the historical data is insufficient, adopt transfer learning technology, based on the similar material-specific prediction sub-model, and adjust the parameters of the temperature-sensitive deep learning model in combination with the mechanical property parameters; When calculating the dynamic compensation roll gap sequence, add a safety margin compensation term for new materials or materials with insufficient data.

[0044] Among them, when integrating a spectral detector in the thickness measurement device of the integrated temperature sensor, when the sheet to be rolled enters the detection area, the spectral detector will collect spectral data on the surface of the sheet in real time. For example, in a rolling workshop, the spectral detector collects spectral information on the surface of the sheet at a frequency of 10 times per second. The collected spectral data will be transmitted to the control system and analyzed through a preset material identification model. This material identification model is trained based on the spectral characteristics of a large number of known materials. It can online identify the material type of the currently rolled sheet, such as carbon steel, aluminum alloy or stainless steel, etc., and at the same time estimate the mechanical property parameters of the material, such as elastic modulus, yield strength, etc. For example, for a sheet to be rolled, the material identification model analyzes the spectral data, identifies its material as 304 stainless steel, and obtains that its elastic modulus is about 193GPa and its yield strength is about 205MPa.

[0045] According to the material type and mechanical property parameters identified by the material identification model, the control system will retrieve the corresponding material-specific prediction sub-model from the pre-established model library. The model library stores the specific prediction sub-models for different common materials. These sub-models are trained based on a large amount of historical rolling data of the material and can more accurately predict the thickness change trend of the material during the rolling process. For example, if the identified material of the sheet is 6061 aluminum alloy, the control system will retrieve the specific prediction sub-model of 6061 aluminum alloy. This sub-model will take real-time temperature data, real-time thickness data and historical rolling condition parameters as inputs and output a prediction matrix of the thickness change trend of the sheet that is more in line with the characteristics of the material.

[0046] In actual production, it may be encountered that the sheet materials of new materials or those with insufficient historical data. When such a situation occurs, transfer learning technology is adopted to adjust the parameters of the temperature-sensitive deep learning model. First, the control system will search for the material-specific prediction submodels similar to the new material in the model library. For example, for a new type of titanium alloy sheet, due to its insufficient historical data, the control system finds that it is relatively similar to the common TC4 titanium alloy in composition and properties, so the material-specific prediction submodel of TC4 titanium alloy is selected as the basis. Then, combined with the mechanical property parameters of the new material identified by spectral data, the parameters of the temperature-sensitive deep learning model are adjusted. Through transfer learning, the model can quickly adapt to the characteristics of the new material. Although the historical data of the new material is limited, it can still accurately predict the thickness change trend during the rolling process to a certain extent.

[0047] When calculating the dynamic compensation roll gap sequence, for new materials or materials with insufficient data, it is necessary to add a safety margin compensation term. This is to reduce the rolling risk caused by insufficient understanding of the characteristics of new materials. For example, for the above-mentioned new type of titanium alloy sheet, when calculating the dynamic compensation roll gap sequence, on the basis of the originally calculated gap adjustment amount, an additional safety margin compensation term is added, such as increasing the overall roll gap by 0.05mm. This can avoid problems such as over-pressing of the sheet or excessive thickness deviation caused by inaccurate prediction of the deformation characteristics of the new material during the rolling process. During the subsequent rolling process, as the data of the new material gradually collected increases, the safety margin compensation term can be dynamically adjusted according to the actual situation to achieve more accurate rolling control.

[0048] Specifically, in some possible embodiments, the method includes: Obtaining the humidity, air pressure and equipment vibration data of the rolling environment in real time through the environmental sensors set at the rolling site; Analyzing the influence laws of the humidity, air pressure and equipment vibration data on the temperature transfer and thickness change during the sheet rolling process, and constructing an environmental correction model; Inputting the humidity, air pressure and equipment vibration data, the real-time temperature data, the real-time thickness data, the historical rolling condition parameters, the material type and the mechanical property parameters into the temperature-sensitive deep learning model or the material-specific prediction submodel; Correcting the prediction matrix of the sheet thickness change trend.

[0049] Among them, in the actual metal sheet rolling production, the influence of environmental factors on sheet thickness control is often complex and diverse. Changes in humidity and air pressure will change the heat dissipation efficiency of materials, and equipment vibration may lead to dynamic fluctuations in the gap between the pressure rollers. In order to accurately capture these influences, a variety of environmental sensors are deployed at key positions on the rolling site: a high-precision temperature and humidity transmitter is installed near the entrance of the rolling mill to collect air humidity and air pressure data in real time. For example, the humidity is monitored at a frequency of 100 times per second to be 65%, and the air pressure is 101.3 kPa; a three-axis vibration acceleration sensor is fixed on the pressure roller bearing block and the frame to capture the amplitude and frequency of equipment vibration in real time. For example, a periodic vibration signal of 150 Hz and 0.3 g is detected. These sensors are connected to the control system in real time through an industrial network to ensure that the environmental data is completely synchronized with the thickness and temperature data of the sheet, providing reliable raw data for subsequent analysis.

[0050] After the environmental data is collected, the system will use the accumulated rolling data and combine machine learning algorithms to deeply analyze the correlation laws between humidity, air pressure, equipment vibration and sheet thickness changes. For example, through a large number of samples, it is found that when the humidity increases by 10%, the heat dissipation rate of the sheet surface to the air will decrease by 5%, resulting in a slowdown in the sheet temperature drop rate of 0.5 °C / second under the same cooling conditions; and when the amplitude of equipment vibration in the 100-200 Hz frequency band increases by 0.1 g, the actual fluctuation range of the pressure roller gap will expand by 0.002 mm, directly affecting the stability of thickness measurement. Based on these laws, the system constructs an environmental correction model, which converts parameters such as humidity, air pressure, and vibration into correction factors for key physical quantities such as temperature transfer efficiency and equipment stiffness. For example, a humidity correction factor K_h = 0.95 is generated, indicating a 5% reduction in heat dissipation efficiency at the current humidity, and a vibration correction factor K_v = 1.02, indicating an equivalent 2% decrease in the stiffness of the pressure roller caused by vibration.

[0051] These environmental correction factors do not act independently, but are input into a temperature-sensitive deep learning model together with real-time temperature, thickness, material parameters, etc. The input layer of the model thus expands the environmental parameter dimension, forming a comprehensive input vector containing more than a dozen features such as material type, mechanical properties, temperature, thickness, humidity, air pressure, and vibration. When calculating the trend of sheet thickness change, the model will automatically call the output results of the environmental correction model to adjust the temperature evolution path and material deformation law. For example, when rolling aluminum alloy sheets in a high-humidity environment, the model will identify the slower heat dissipation caused by humidity, thereby prolonging the duration of the sheet in the high-temperature state and correspondingly adjusting the prediction of the decrease in the yield strength of the material, making the calculation of the influence of temperature on thickness change closer to the actual working conditions.

[0052] The incorporation of environmental data directly affects the generation of the dynamic compensation roll gap sequence. When the environmental correction model detects abnormal equipment vibration or significant changes in humidity, the system will add targeted compensation measures to the original gap adjustment strategy. For example, when the vibration acceleration exceeds 0.5g, in order to offset the risk of gap fluctuations caused by vibration, an additional safety margin of 0.003mm will be added to the calculated gap adjustment amount; when the humidity is 20% higher than the reference value, considering that the decrease in heat dissipation efficiency may lead to a higher temperature of the sheet material and increased softening of the material, the gap adjustment amount in the middle area of the roll will be reduced in advance to avoid excessive thinning due to the enhanced deformation ability of the material at high temperatures. This mechanism that converts environmental impacts into control strategies in real time enables the system to actively adapt to the subtle changes in the workshop environment rather than relying on fixed empirical parameters.

[0053] Taking the rolling process of a certain stainless steel sheet in a high-humidity environment as an example, when the sensor detects that the humidity reaches 75% (15% higher than the reference value), the air pressure slightly decreases, and the equipment vibration is at a medium level, the environmental correction model first calculates that the heat dissipation efficiency decreases by 12%. This means that the cooling rate of the sheet during rolling will be slower than the ideal state, and the temperature may be 3°C higher than expected. The temperature-sensitive model adjusts the thickness change prediction accordingly, and additionally considers a 0.008mm thinning amount caused by the decrease in the yield strength of the material due to high temperature in the original calculation. Subsequently, in the generated dynamic compensation gap sequence, the gap adjustment amount in the middle area is reduced from the initial 0.05mm to 0.042mm to offset the excessive thinning caused by high temperature, and at the same time, a global safety margin of 0.003mm is added for vibration risk. The final measured results show that the thickness deviation of the sheet is significantly reduced from ±0.03mm of the traditional method to ±0.015mm, fully demonstrating the actual effect of environmental factor correction on improving rolling accuracy. By deeply integrating environmental data into the control process in this way, the rolling system is no longer limited by fixed environmental assumptions, but can dynamically adjust the strategy according to the real-time changing workshop conditions, fundamentally enhancing its adaptability to complex production environments.

[0054] Specifically, in some possible embodiments, the method includes: When the temperature change rate of the sheet is monitored to exceed the threshold value, based on the material type, mechanical property parameters, target thickness of the current sheet, and the humidity, air pressure, and equipment vibration data, calculate the optimal stable temperature range through the temperature-sensitive deep learning model; Adjust the temperature of the sheet to the optimal stable temperature range through the heating device or cooling device integrated on the rolling line, and the adjustment power of the heating device or cooling device is dynamically matched according to the real-time heat radiation data of the sheet; Synchronously input the adjusted temperature data after stabilizing the sheet temperature, the material type, the mechanical property parameters, the humidity, the air pressure, and the equipment vibration data into the temperature-sensitive deep learning model to generate a low-fluctuation exclusive prediction matrix; Calculate the dynamic compensation roll gap sequence based on the low-fluctuation exclusive prediction matrix.

[0055] Among them, during the actual rolling process of metal sheets, drastic temperature changes will significantly affect the mechanical properties of the material, thereby increasing the difficulty of thickness control. When the system real-time monitors through the integrated temperature sensor that the temperature change rate of the sheet exceeds the preset threshold (for example, the temperature rise or fall per second exceeds 10 °C), a targeted temperature stabilization control mechanism will be triggered. Taking the hot rolling process of a certain titanium alloy sheet for aerospace (target thickness 1.5 mm) as an example, when it is detected that the surface temperature of the sheet rises suddenly from 800 °C to 850 °C (change rate 100 °C / s) within 50 ms, far exceeding the preset safety threshold of 20 °C / s, the system immediately starts the following process: First of all, the temperature-sensitive deep learning model will combine the material type of the current sheet (titanium alloy TC4), the mechanical property parameters (thermal expansion coefficient 8.6×10⁻ 6 / °C, yield strength 850 MPa), the target thickness, and the real-time environmental data (humidity 60%, air pressure 101 kPa, equipment vibration acceleration 0.2 g) to calculate the optimal stable temperature range suitable for this material. The model analyzes historical data and finds that within the temperature range of 820 - 840 °C for TC4 titanium alloy, the thermal expansion coefficient fluctuation is less than 5%, and the yield strength change rate is lower than 3%. This is the temperature range where the deformation resistance of the material is the most stable during the rolling process. Therefore, the system sets 820 - 840 °C as the optimal stable temperature range under the current working conditions to avoid problems such as increased material softening and uneven thermal expansion caused by continuous rapid temperature rise.

[0056] Next, the heating / cooling device integrated on the rolling line will dynamically adjust the power according to the real-time thermal radiation data of the sheet. The infrared thermal imager installed above the rolling mill scans the surface temperature distribution of the sheet in real time and finds that the current temperature in the middle of the sheet has reached 860 °C, and the edge temperature is 830 °C, with an obvious temperature gradient. The system calculates the required adjustment power for each region accordingly: start the high-pressure air cooling device for the high-temperature region in the middle, increase the cooling air volume from 500 m³ / h to 800 m³ / h, so that the temperature in the middle drops at a rate of 20 °C / s; turn on the infrared heating lamp for the low-temperature region at the edge, increase the power from 20 kW to 30 kW, and ensure that the edge temperature is maintained at 830 °C. Through this dynamic power matching, the overall temperature of the sheet is stabilized within the range of 825 - 835 °C within 200 ms, meeting the requirements of the optimal stable temperature.

[0057] After the temperature of the sheet material is adjusted and stabilized, the new adjusted temperature data (such as 830°C in the middle and 825°C at the edges) will be synchronously input into the temperature-sensitive deep learning model together with the material parameters and environmental data. At this time, the "low-fluctuation exclusive prediction matrix" generated by the model is no longer disturbed by sudden temperature changes and can more accurately reflect the deformation law of the material under a stable thermal state. For example, the model predicts that within the stable temperature range, the difference in thermal expansion of each area of the sheet material will be reduced from 0.05 mm during sudden changes to less than 0.01 mm, and the risk of thickness change caused by the yield strength fluctuation is reduced by 60%. This prediction matrix based on stable temperature provides a more reliable basis for subsequent gap adjustment.

[0058] Finally, when the system calculates the dynamic compensation roll gap sequence based on the low-fluctuation exclusive prediction matrix, there is no need to additionally consider the uncertainty brought by drastic temperature changes. For example, for the slight thickening trend (predicted thickening of 0.015 mm) in the edge area of the titanium alloy sheet at a stable temperature, the system only needs to reduce the edge gap of the roll by 0.01 mm to accurately compensate for the thickness deviation, avoiding the problems of over-adjustment or adjustment lag caused by temperature fluctuations in the traditional method. The measured data shows that after adopting this mechanism, the thickness deviation of the titanium alloy sheet under the condition of sudden temperature change is stably controlled within ±0.01 mm from ±0.04 mm, effectively solving the problem of uneven deformation caused by unstable temperature during high-speed rolling of heat-sensitive materials, and improving the yield rate and reliability of high-end sheet materials.

[0059] Specifically, in some possible embodiments, the method includes: When the rolled sheet material is a dissimilar metal composite sheet, an ultrasonic flaw detector and a metallographic microscope probe are added to the thickness measuring device with an integrated temperature sensor to obtain the bonding condition data of the interface of the dissimilar metal composite sheet in real time. The bonding condition data includes bonding strength information and interface defect information; Spectral analysis is performed on each layer of metal of the dissimilar metal composite sheet, and the material type and mechanical property parameters of each layer of metal are accurately identified through a preset material identification model; At the same time, a comprehensive mechanical property model of the dissimilar metal composite sheet is established, considering the interaction between each layer of metal to more accurately describe the overall mechanical behavior of the dissimilar metal composite sheet; According to the material type, the mechanical property parameters of each layer of metal, and the bonding condition data, a customized temperature-sensitive deep learning sub-model is developed. The temperature-sensitive deep learning sub-model focuses on the heat transfer and deformation coordination relationship between different metal layers during the training process to improve the prediction accuracy of the thickness change trend of the dissimilar metal composite sheet; When calculating the dynamic compensation roll gap sequence, the target thickness of the sheet and the mechanical characteristic parameters of the rolls are considered, and in combination with the combined situation data and the comprehensive mechanical property model of the dissimilar metal composite sheet, an additional gap compensation amount is added for the weak interface bonding area to ensure uniform deformation and good bonding of each layer of metal in the dissimilar metal composite sheet during the rolling process.

[0060] Among them, when actually rolling dissimilar metal composite sheets (such as aluminum / steel composite plates, titanium / copper composite plates), for the special requirements of interlayer bonding and deformation coordination, targeted upgrades of hardware and algorithms are required on the basis of existing detection devices. An ultrasonic flaw detector and a metallographic microscope probe are added to the thickness measuring device integrated with temperature sensors. The ultrasonic flaw detector scans the sheet interface at a frequency of 50 times per second, and analyzes the bonding strength and identifies interface defects (such as cracks and inclusions, with a minimum detection size of 50 μm) through the reflected echo signal; the metallographic microscope probe collects the microscopic structure images of the interface in real time to detect the interlayer metallurgical bonding state (such as the degree of grain interlocking). When the bonding strength of a certain area is detected to be only 80 MPa (lower than the critical value) or there are microcracks, the system immediately marks it as a "weak bonding area" to provide accurate positioning for subsequent control.

[0061] Secondly, the chemical compositions of the upper and lower layers of metal of the composite sheet are analyzed separately by a spectral detector. The upper aluminum layer uses 3003 aluminum alloy (thermal expansion coefficient of 23×10⁻ 6 / °C, yield strength of 110 MPa), and the lower steel layer is Q235 (thermal expansion coefficient of 12×10⁻ 6 / °C, yield strength of 235 MPa). The material identification model independently outputs mechanical parameters based on the spectral data of each layer and establishes a comprehensive mechanical property model, and simulates the interlayer interaction by the finite element method: when the temperature of the aluminum layer is 20 °C higher than that of the steel layer, the thermal expansion difference will generate a shear stress of 5 MPa at the interface. If the bonding strength is insufficient, interlayer sliding may occur. The model converts this interlayer stress-strain relationship into an overall deformation correction coefficient. For example, the actual reduction rate of the aluminum layer needs to consider the constraint effect of the steel layer, and the correction coefficient is 0.85.

[0062] Based on the above data, the system customizes a temperature-sensitive deep learning sub-model. In addition to the traditional input parameters such as temperature, thickness, and working conditions, new input parameters are added, including the interfacial bonding strength (e.g., 100 MPa), the defect location (e.g., 100 mm from the edge), and the material parameters of each layer (the elastic modulus of the aluminum layer is 70 GPa, and that of the steel layer is 210 GPa). During model training, key learning focuses on the interlayer heat transfer delay (the measured time constant for heat conduction from the aluminum layer to the steel layer is 20 ms) and the deformation coordination law (e.g., for every 0.01 mm reduction in the thickness of the aluminum layer, the steel layer reduces by 0.006 mm due to the constraint effect). The output thickness change trend prediction matrix is no longer a global prediction but is layer-distinguished (e.g., the aluminum layer is predicted to thin by 0.02 mm, and the steel layer is predicted to thin by 0.015 mm), and the interfacial stress concentration areas are marked (e.g., the risk of stress exceeding the critical value is high within 20 mm of the edge).

[0063] When calculating the dynamic compensation roll gap sequence, the system first calculates the basic gap adjustment amount (e.g., the overall gap is reduced by 0.1 mm) according to the target thickness (total thickness 2.0 mm, aluminum layer 0.45 mm, steel layer 1.55 mm) and the mechanical characteristics of the rolls (the roll body stiffness in the middle is greater than that at both ends), in combination with the comprehensive mechanical model. For the weak interfacial bonding areas (e.g., the marked 100 mm from the edge), the system automatically increases the additional compensation amount: if the bonding strength is 80 MPa (80% of the critical value), then on the basis of the gap adjustment amount in this area, it is increased by 0.005 mm to reduce the interlayer shear stress (from 5 MPa to 3 MPa). This zoning compensation strategy balances the differences in the deformation resistance of each layer of metal (the uneven reduction rate caused by the soft aluminum and hard steel) through basic adjustment and enhancement of weak areas, avoiding tearing or peeling of the bonding surface caused by uniform gap adjustment.

[0064] During the actual rolling process, when it is detected that the interfacial bonding strength of a certain batch of composite plates is generally low (average 90 MPa), the sub-model will automatically increase the additional compensation amount in the edge area from 0.005 mm to 0.01 mm and extend the low-speed stable rolling time of the rolls (from 10 s to 15 s) to ensure the uniformity of interlayer deformation, significantly improving the manufacturing quality of dissimilar metal composite plates.

[0065] Specifically, in some possible embodiments, the method includes: Real-time obtaining the temperature change data of each layer of metal through multiple high-precision temperature sensors set at the rolling site; Calculating the thermal expansion amounts at different positions according to the thermal expansion coefficients of each layer of metal to generate a thermal expansion distribution map; Determining the metal layers that need to be temperature-controlled and their control directions according to the differences in the thermal expansion amounts of each layer of metal in the thermal expansion distribution map; The heating device or cooling device integrated through the rolling line independently controls the temperature of each metal layer; Monitor the change in thermal expansion of each metal layer in real time, and adjust the heating or cooling rate according to the change in thermal expansion; Obtain the thermal expansion monitoring data of each metal layer; Obtain the temperature control parameters of the heating device or cooling device; Input the thermal expansion monitoring data and the temperature control parameters into the temperature-sensitive deep learning sub-model to update the prediction matrix of the change trend of the sheet thickness in real time, so as to more accurately reflect the influence of thermal expansion differences on the thickness and interface bonding of the dissimilar metal composite sheet.

[0066] Among them, at different key positions of the rolling mill at the rolling site, such as the entrance, the rolling area, and the exit, a plurality of high-precision temperature sensors are installed. For the aluminum-steel composite plate, sensors are respectively arranged at corresponding positions of the aluminum plate layer and the steel plate layer. These sensors have the characteristics of high precision and fast response, and can collect the temperature data of each metal layer in real time at a frequency of 10 times per second. For example, at the beginning of rolling, the temperature of the aluminum plate layer is 200 °C, and the temperature of the steel plate layer is 150 °C.

[0067] The thermal expansion coefficients of the aluminum plate and the steel plate are known. According to the collected temperature change data of each metal layer, combined with the thermal expansion coefficient formula ΔL = L0×α×ΔT (where ΔL is the thermal expansion, L0 is the initial length, α is the thermal expansion coefficient, and ΔT is the temperature change), calculate the thermal expansion at different positions. Assume that the initial lengths of the aluminum plate and the steel plate are both 1 m. When the temperature of the aluminum plate rises by 50 °C and the temperature of the steel plate rises by 30 °C, the calculated thermal expansion of the aluminum plate is 0.00115 m, and the thermal expansion of the steel plate is 0.00036 m. By calculating the thermal expansion at different positions of the entire sheet, a thermal expansion distribution map is generated to visually display the difference in thermal expansion of each metal layer.

[0068] Analyze the thermal expansion distribution map. When it is found that the difference in thermal expansion between the aluminum plate layer and the steel plate layer is large, temperature control is required. If the thermal expansion of the aluminum plate is too large, it may cause problems such as warping and poor interface bonding of the composite plate. At this time, it is determined that the temperature of the aluminum plate layer needs to be controlled, and the control direction is to lower the temperature; conversely, if the thermal expansion of the steel plate is too small, the control direction is to raise the temperature.

[0069] The rolling line is integrated with a heating device and a cooling device. For the aluminum plate layers that need to be cooled, start the cooling device, such as by using air cooling or water cooling. By adjusting the flow rate and temperature of the cooling medium, precisely control the cooling rate of the aluminum plate layers. For example, turn on the air cooling device and set the flow rate of the cooling air to 10 m³ / min and the temperature to 20 °C. For the steel plate layers that need to be heated, start the heating device, such as by using induction heating or resistance heating. By adjusting the heating power, control the heating rate of the steel plate layers.

[0070] During the temperature control process, continuously and real-time monitor the changes in the thermal expansion of each metal layer. Devices such as displacement sensors can be used to measure the length changes of each layer of metal at a frequency of 5 times per second, so as to calculate the real-time changes in the thermal expansion. According to the changes in the thermal expansion, dynamically adjust the heating or cooling rate. If it is found that the rate of decrease in the thermal expansion of the aluminum plate layer is too slow, it means that the cooling rate is insufficient, and the flow rate of the cooling air can be increased to 15 m³ / min; if the rate of increase in the thermal expansion of the steel plate layer is too fast, the heating power can be reduced to 40 kW.

[0071] Throughout the process, continuously obtain the thermal expansion monitoring data of each metal layer, including the real-time values of the thermal expansion, the change rate, etc. At the same time, record the temperature control parameters of the heating device or the cooling device, such as the heating power, the flow rate and temperature of the cooling medium, etc. Input these thermal expansion monitoring data and temperature control parameters into the temperature-sensitive deep learning sub-model. This sub-model will update the prediction matrix of the change trend of the plate thickness in real time according to the newly input data. For example, originally it was predicted that the overall thickness of the composite plate would increase by 0.1 mm. Considering the real-time changes in the thermal expansion difference, the predicted thickness is adjusted to an increase of 0.08 mm, so as to more accurately reflect the influence of the thermal expansion difference on the thickness and interface bonding of the dissimilar metal composite plate. It can effectively control and manage the thermal expansion differences of each layer of metal in the dissimilar metal composite plate, and improve the stability of the rolling process and the quality of the composite plate.

[0072] Specifically, in some possible embodiments, the method includes: Embed distributed fiber Bragg grating sensors inside the roll body and bearing housing of the pressure roll to collect the temperature distribution data of the circumferential and axial directions of the pressure roll in real time, and obtain the surface temperature gradient of the pressure roll and the temperature rise rate of the bearing housing; Combined with the thermal expansion coefficient of the roll material, construct a thermal deformation prediction model of the roll, and calculate the clearance deviation caused by the radial expansion of the roll due to temperature change based on the heat conduction equation and the expansion formula; Input the temperature distribution data, the clearance deviation, the real-time thickness data and the real-time temperature data into the temperature-sensitive deep learning model to generate a thermal compensation dynamic clearance sequence; Collect the thickness feedback data of the rolled sheet through the thickness measuring device. If the thickness deviation exceeds the preset range within a continuous preset period and the deviation direction is consistent with the predicted direction of the hot deformation of the press roll, extract the actual temperature of the press roll, the actual gap between the press roll and the sheet, and the actual thickness of the sheet at the current moment, and update the hot deformation weight parameters of the temperature-sensitive deep learning model to achieve the adaptive optimization of the compensation strategy.

[0073] Among them, during the actual operation of the rolling mill equipment, the hot deformation of the press roll will cause gap deviation, which in turn affects the thickness accuracy of the sheet rolling. Install distributed fiber Bragg grating sensors inside the press roll body and the bearing housing. On the press roll body, install sensors at regular intervals along the circumferential direction and at regular distances along the axial direction; also arrange sensors at positions close to the outer ring of the bearing inside the bearing housing. These sensors can collect data with high precision and quickly, such as collecting 100 times of data per second. When the press roll rotates at high speed, the sensors will maintain real-time communication with a specific device and control system to ensure that the circumferential and axial temperature distributions of the press roll can be accurately obtained. For example, it can be monitored that the temperature in the middle of the press roll body is higher than that at both ends, and it can also be known the heating rate of the bearing housing due to bearing friction.

[0074] Using the collected temperature data and combining with the thermal expansion characteristics of the press roll material, construct a model that can predict the hot deformation of the press roll. This model is based on the principle of heat conduction. By simulating the temperature change inside the press roll, calculate the expansion degree of the press roll in the radial direction due to temperature change. According to the expansion degree of the press roll, the deviation value of the gap between the press roll and the sheet can be further obtained. For example, it can be calculated how much the gap in a certain area of the press roll is reduced due to thermal expansion, so as to obtain the gap deviation conditions at different axial positions of the entire press roll.

[0075] Input the temperature distribution data of the press roll, the calculated gap deviation data, as well as the real-time thickness and temperature data of the sheet into the temperature-sensitive deep learning model. This model will analyze and process these data to find the correlation between the hot deformation of the press roll and the thickness deviation of the sheet. For example, it is judged that the thermal expansion in a certain area of the press roll is the main reason for the thickness deviation in the corresponding area of the sheet. Then, the model will generate a dynamic gap sequence for thermal compensation, which contains the adjustment strategies for the gaps at different positions of the press roll in the next period of time. Through the electro-hydraulic servo system, the position of the press roll can be quickly adjusted according to this sequence to dynamically correct the gap.

[0076] After the sheet rolling is completed, thickness feedback data of the sheet is collected by a thickness measuring device. If the thickness deviation of the sheet measured in several consecutive cycles exceeds a pre-set range, and the direction of the deviation is consistent with the direction predicted by the previous thermal deformation of the press roll, it indicates that there may be an error in the model. At this time, the system will extract data such as the actual temperature of the press roll at the current moment, the actual gap between the press roll and the sheet, and the actual thickness of the sheet. Using these data, the weight parameters regarding the thermal deformation of the press roll in the temperature-sensitive deep learning model are updated through a specific algorithm. After multiple iterative adjustments, the model can more accurately predict the impact of the thermal deformation of the press roll on the gap, and control the thickness deviation of the sheet within a smaller range.

[0077] During the high-speed rolling of stainless steel sheets, a large amount of heat is generated due to the friction between the press roll and the sheet, resulting in a rapid increase in the temperature of the bearing housing and a significant increase in the temperature in the middle of the roll body. If thermal deformation compensation is not carried out, the thickness deviation of the sheet will become larger and larger. After adopting the above solution, the sensor can timely capture the thermal expansion of the press roll, the model will generate a corresponding thermal compensation gap sequence to adjust the press roll gap, and at the same time, continuously optimize the model parameters through feedback calibration. Finally, the thickness deviation of the sheet can be controlled within a very small range, ensuring the quality stability of the sheet during the high-speed rolling process.

[0078] This method that combines equipment thermal state monitoring, model calculation, and data-driven prediction forms a complete closed-loop control system. Through accurate temperature data acquisition, prediction of the thermal deformation model, and adaptive adjustment of the deep learning model, the system can respond in real time to the impact brought by the thermal deformation of the press roll, effectively improving the accuracy and stability of sheet rolling, especially suitable for high-speed and high-temperature rolling production scenarios.

[0079] Another embodiment of this application provides a rolling system in a fully closed mode, wherein, refer to Figure 2 , a rolling system in a fully closed mode, including: Thickness and temperature data acquisition module 100: Real-time thickness data of the sheet to be rolled and real-time temperature data at the corresponding position are obtained in real time through a thickness measuring device integrated with a temperature sensor; Sheet thickness change trend prediction matrix output module 200: A temperature-sensitive deep learning model is preset, and using the real-time temperature data, the real-time thickness data, and historical rolling condition parameters as inputs, a sheet thickness change trend prediction matrix under different temperature evolution paths within a future preset time is output; Dynamic compensation press roll gap sequence calculation module 300: Based on the sheet thickness change trend prediction matrix, combined with the target thickness of the sheet and the current mechanical characteristics parameters of the press roll, a dynamic compensation press roll gap sequence is calculated, and the dynamic compensation press roll gap sequence includes the gap adjustment amounts and adjustment time sequences for several future control cycles; Roller adjustment and feedback data upload module 400: Send the dynamic compensation roller gap sequence to the roller control system. The roller control system dynamically adjusts the roller gap, synchronously collects the thickness feedback data and the corresponding temperature feedback data of the sheet after the roller adjustment and rolling is completed, and uploads the thickness feedback data and the temperature feedback data.

[0080] For the rolling system in the fully closed mode provided in this embodiment, due to the functions of its respective modules and the logical connections between them, it can implement each step of the foregoing embodiment, so it can achieve the same technical effects as the foregoing embodiment. For the principle analysis, reference can be made to the relevant descriptions of the steps of the foregoing rolling method in the fully closed mode, which will not be repeated here.

[0081] An embodiment of the present application also provides a rolling device in the fully closed mode, including a memory and a processor. A computer program capable of being loaded and executed by the processor for the foregoing rolling method in the fully closed mode is stored on the memory.

[0082] An embodiment of the present application also provides a storage medium, in which a computer program capable of being loaded and executed by the processor for the foregoing rolling method in the fully closed mode is stored.

[0083] For the storage medium provided in this embodiment, since the computer program therein, after being loaded and run on the processor, will implement each step of the foregoing embodiment, it can achieve the same technical effects as the foregoing embodiment. For the principle analysis, reference can be made to the relevant descriptions of the foregoing method steps, which will not be repeated here.

[0084] The storage medium includes, for example: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0085] The steps of the method or algorithm described in combination with the embodiments disclosed in this article can be directly implemented by hardware, software modules executed by a processor, or a combination of both. The software module can be placed in a random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, or any other form of storage medium well known in the technical field.

[0086] In the description of this specification, the descriptions with reference to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0087] In addition, features defined by terms "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, the meaning of "a plurality of" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. It is only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features.

[0088] Thus, any process or method description shown in the flowchart or described in other ways herein can be understood to represent a module, segment, or part of code including one or more executable instructions for implementing a customized logic function or process. And the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0089] The embodiments of this specific implementation manner are all preferred embodiments of this application, and do not limit the protection scope of this application. Therefore, all equivalent changes made according to the structure, shape, and principle of this application shall be covered by the protection scope of this application.

Claims

1. A rolling method in a fully closed mode, characterized in that, Including: Real-time thickness data of the to-be-rolled sheet and real-time temperature data at the corresponding position are obtained in real time by a thickness measurement device integrating a temperature sensor; A temperature-sensitive deep learning model is preset, and the real-time temperature data, the real-time thickness data, and historical rolling working condition parameters are used as inputs to output a prediction matrix of the thickness change trend of the sheet under different temperature evolution paths within a preset future time; Based on the prediction matrix of the sheet thickness change trend, combined with the target thickness of the sheet and the current mechanical characteristics parameters of the pressure roller, a dynamic compensation pressure roller gap sequence is calculated, and the dynamic compensation pressure roller gap sequence includes the gap adjustment amounts and adjustment time sequences for several future control cycles; The dynamic compensation pressure roller gap sequence is sent to the pressure roller control system, and the pressure roller control system dynamically adjusts the pressure roller gap, synchronously collects the thickness feedback data and the corresponding temperature feedback data of the sheet after the pressure roller adjustment and completed rolling, and uploads the thickness feedback data and the temperature feedback data.

2. The rolling method in a fully closed mode according to claim 1, characterized in that The method includes: A spectral detector is integrated in the thickness measurement device integrating the temperature sensor to collect the surface spectral data of the sheet in real time, and the material type and mechanical property parameters of the currently rolled sheet are identified online through a preset material identification model; According to the identified material type and mechanical property parameters, a corresponding material-specific prediction sub-model is retrieved; If the sheet is a new material and the historical data is insufficient, transfer learning technology is used to adjust the parameters of the temperature-sensitive deep learning model based on a similar material-specific prediction sub-model and combined with the mechanical property parameters; When calculating the dynamic compensation pressure roller gap sequence, a safety margin compensation term is added for new materials or materials with insufficient data.

3. A rolling method in a fully closed mode according to claim 2, characterized in that, The method includes: The humidity, air pressure, and equipment vibration data of the rolling environment are obtained in real time through environmental sensors set at the rolling site; The influence laws of the humidity, air pressure, and equipment vibration data on the temperature transfer and thickness change during the sheet rolling process are analyzed, and an environment correction model is constructed; The humidity, air pressure, and equipment vibration data, the real-time temperature data, the real-time thickness data, the historical rolling working condition parameters, the material type, and the mechanical property parameters are input into the temperature-sensitive deep learning model or the material-specific prediction sub-model; The prediction matrix of the sheet thickness change trend is corrected.

4. A rolling method in a fully closed mode according to claim 3, characterized in that The method includes: When it is monitored that the temperature change rate of the sheet exceeds a threshold value, based on the material type, the mechanical property parameters, the target thickness, and the humidity, air pressure, and equipment vibration data of the current sheet, the optimal stable temperature range is calculated through the temperature-sensitive deep learning model; The temperature of the sheet is adjusted to the optimal stable temperature range through a heating device or a cooling device integrated on the rolling line, and the adjustment power of the heating device or the cooling device is dynamically matched according to the real-time thermal radiation data of the sheet; The adjusted temperature data, the material type, the mechanical property parameters, the humidity, air pressure, and equipment vibration data after the sheet temperature is adjusted stably are synchronously input into the temperature-sensitive deep learning model to generate a low-fluctuation specific prediction matrix; Calculate the dynamic compensation roll gap sequence based on the low-fluctuation exclusive prediction matrix.

5. A rolling method in a fully closed mode according to claim 2, characterized in that The method includes: When the rolled sheet is a dissimilar metal composite sheet, an ultrasonic flaw detector and a metallographic microscope probe are added to the thickness measuring device of the integrated temperature sensor to obtain the bonding condition data of the interface of the dissimilar metal composite sheet in real time. The bonding condition data includes bonding strength information and interface defect information; Perform spectral analysis on each layer of metal of the dissimilar metal composite sheet, and accurately identify the material type and mechanical property parameters of each layer of metal through a preset material identification model; At the same time, establish a comprehensive mechanical property model of the dissimilar metal composite sheet, considering the interaction between each layer of metal to more accurately describe the overall mechanical behavior of the dissimilar metal composite sheet; Customize a dedicated temperature-sensitive deep learning sub-model according to the material type, mechanical property parameters, and bonding condition data of each layer of metal. The temperature-sensitive deep learning sub-model focuses on the heat transfer and deformation coordination relationship between different metal layers during the training process to improve the prediction accuracy of the thickness change trend of the dissimilar metal composite sheet; When calculating the dynamic compensation roll gap sequence, consider the target thickness of the sheet and the roll mechanical characteristic parameters, and combine the bonding condition data and the comprehensive mechanical property model of the dissimilar metal composite sheet. For the weak bonding area of the interface, an additional gap compensation amount is added to ensure uniform deformation and good bonding of each layer of metal of the dissimilar metal composite sheet during the rolling process.

6. A rolling method in a fully closed mode according to claim 5, characterized in that The method includes: Obtain the temperature change data of each layer of metal in real time through multiple high-precision temperature sensors set at the rolling site; Calculate the thermal expansion amount at different positions according to the thermal expansion coefficient of each layer of metal to generate a thermal expansion distribution map; Determine the metal layer that needs to be temperature-controlled and its control direction according to the thermal expansion amount difference of each layer of metal in the thermal expansion distribution map; Independently control the temperature of each layer of metal through the heating device or cooling device integrated on the rolling line; Monitor the change of the thermal expansion amount of each metal layer in real time, and adjust the heating or cooling rate according to the change of the thermal expansion amount; Obtain the thermal expansion monitoring data of each metal layer; Obtain the temperature control parameters of the heating device or cooling device; Input the thermal expansion monitoring data and the temperature control parameters into the temperature-sensitive deep learning sub-model to update the thickness change trend prediction matrix of the sheet in real time to more accurately reflect the influence of thermal expansion difference on the thickness and interface bonding of the dissimilar metal composite sheet.

7. A rolling method in a fully closed mode according to claim 1, characterized in that, The method includes: Embed distributed fiber Bragg grating sensors inside the roll body and bearing housing of the roll to collect the temperature distribution data in the circumferential and axial directions of the roll in real time, and obtain the surface temperature gradient of the roll and the temperature rise rate of the bearing housing; Combine the thermal expansion coefficient of the roll material to construct a roll thermal deformation prediction model, and calculate the gap deviation caused by the radial expansion of the roll due to temperature change based on the heat conduction equation and the expansion formula. Input the temperature distribution data, the gap deviation, the real-time thickness data, and the real-time temperature data into the temperature-sensitive deep learning model to generate a thermally compensated dynamic gap sequence; Collect the thickness feedback data of the rolled sheet by the thickness measuring device. If the thickness deviation exceeds the preset range within a continuous preset period and the deviation direction is consistent with the predicted direction of the thermal deformation of the press roll, extract the actual temperature of the press roll, the actual gap between the press roll and the sheet, and the actual thickness of the sheet at the current moment, and update the thermal deformation weight parameters of the press roll of the temperature-sensitive deep learning model to achieve adaptive optimization of the compensation strategy.

8. A rolling system in a fully closed mode, characterized in that, including: Thickness and temperature data acquisition module: Real-time acquire the real-time thickness data of the sheet to be rolled and the real-time temperature data at the corresponding position through a thickness measuring device integrated with a temperature sensor; Sheet thickness change trend prediction matrix output module: Preset a temperature-sensitive deep learning model, use the real-time temperature data, the real-time thickness data, and the historical rolling condition parameters as inputs, and output a sheet thickness change trend prediction matrix under different temperature evolution paths within a preset future time; Dynamic compensation press roll gap sequence calculation module: Based on the sheet thickness change trend prediction matrix, combined with the target thickness of the sheet and the current press roll mechanical characteristic parameters, calculate a dynamic compensation press roll gap sequence, and the dynamic compensation press roll gap sequence includes the gap adjustment amounts and adjustment time sequences for several future control cycles; Press roll adjustment and feedback data upload module: Send the dynamic compensation press roll gap sequence to the press roll control system, the press roll control system dynamically adjusts the press roll gap, synchronously collects the thickness feedback data and the corresponding temperature feedback data of the sheet after the press roll adjustment and completed rolling, and uploads the thickness feedback data and the temperature feedback data.

9. A rolling equipment in a fully closed mode, characterized in that, including: A memory and a processor, and a computer program capable of being loaded and executed by the processor for the rolling method in any one of the above claims 1-7 in the fully closed mode is stored on the memory.

10. A storage medium, characterized in that, Stored with a computer program capable of being loaded and executed by the processor for the rolling method in any one of the above claims 1-7 in the fully closed mode.

Images