A fully automated production system and method for rolling steel tubes for gravure printing plates

Abstract

This application discloses a fully automated production system and method for rolling steel pipes for gravure printing plates, belonging to the field of steel structure fabrication technology. The system includes: a process requirements analysis module to acquire steel plate process information and derive preliminary process parameters; a process parameter determination module to optimize preliminary parameters based on historical data; a deviation determination module to extract morphological deviation and error information through real-time data analysis; an adjustment module to adjust process parameters according to the deviations; a process parameter correction module to correct process parameters through response data; and a rolling adjustment module to generate a process control strategy based on the optimized parameters and adjust the rolling machine. This solution optimizes rolling process parameters through multi-dimensional data analysis and real-time feedback. By combining historical process data and real-time morphological data, process parameters are dynamically adjusted to ensure the quality accuracy and production stability of the steel pipes, improving the adaptability and efficiency of the rolling process.

Description

Technical Field

[0001] This application belongs to the field of steel structure fabrication technology, specifically relating to a fully automated production system and method for rolling steel tubes for gravure printing plates. Background Technology

[0002] The rolling process of gravure printing plate is a core step in controlling plate-making precision, as its morphological accuracy directly affects the uniformity and stability of the printing roller. Traditional processes are insufficient to meet the requirements of dynamic adjustment and real-time optimization. Therefore, there is an urgent need for data-driven process control methods to achieve precise control throughout the entire process, from initial parameter derivation to closed-loop production feedback, in order to improve production efficiency and finished product quality.

[0003] Currently, initial parameters are usually estimated using empirical formulas, and historical production data is manually recorded and the average is simply calculated as a reference. During the production process, the rolling operation is performed using preset parameters throughout, and the steel plate shape data is only obtained by offline caliper measurement after all processes are completed. If the test results do not meet the standards, the machine must be stopped and the parameters must be adjusted by technicians based on experience. Material condition assessment relies on periodic sampling for destructive testing, and parameter correction is gradually approached to the optimal value through repeated trial production.

[0004] Existing technologies lack real-time data fusion and dynamic feedback mechanisms, and parameter optimization lags behind the production process, leading to the accumulation of morphological deviations. Furthermore, offline detection and manual adjustment are inefficient and difficult to cope with dynamic factors such as material fluctuations. At the same time, destructive sampling limits the frequency of material evaluation, and the trial-and-error correction cycle is long. Summary of the Invention

[0005] To overcome the above-mentioned shortcomings, this invention is proposed to provide solutions or at least partially solve the technical problems of existing technologies, such as the lack of real-time data fusion and dynamic feedback mechanisms, parameter optimization lagging behind the production process, resulting in the accumulation of morphological deviations; the low efficiency of offline detection and manual adjustment, making it difficult to cope with dynamic factors such as material fluctuations; and the limitation of material evaluation frequency by destructive sampling, and the long trial-and-error correction cycle.

[0006] In a first aspect, the present invention provides a fully automated production system for rolling steel tubes for gravure printing plates, the system comprising: The process requirements analysis module is used to obtain the process requirements information of the target gravure printing plate steel plate, perform multi-dimensional data analysis and parameter derivation on the process requirements information, and obtain the preliminary process control parameter set of the target gravure printing plate steel plate. The process parameter determination module is used to acquire historical process data that matches the feature parameters of the target gravure printing plate steel plate, perform time-series feature extraction and multi-dimensional data fusion based on the historical process data, extract sensitive key control variables in the steel plate rolling process, and optimize the preliminary process control parameter set based on the sensitive key control variables to obtain the rolling process parameter set. The deviation determination module is used to control the rolling machine to perform rolling operation on the target gravure printing plate steel sheet based on the rolling process parameter set, obtain the real-time shape data of the target gravure printing plate steel sheet during the rolling process, identify the current process based on the real-time shape data, and perform error analysis and deviation extraction on the real-time shape data and the theoretical forming trajectory reference data corresponding to the current process to obtain the shape deviation data and process execution error information of the target gravure printing plate steel sheet during the rolling process. The adjustment module is used to adjust the rolling process parameter set based on the shape deviation data and process execution error information if the shape deviation data exceeds the preset accuracy standard corresponding to the current process. The process parameter correction module is used to acquire the force and displacement response data of the target gravure printing plate steel plate, invert the material state data of the target gravure printing plate steel plate based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set. The rolling adjustment module is used to acquire process control target data, perform multi-objective optimization calculations based on the process control target data and the corrected rolling process parameter set, generate target process control strategies, and adjust the rolling machine based on the target process control strategies.

[0007] In a second aspect, the present invention provides a fully automated production method for rolling steel tubes for gravure printing plates, the method comprising: Obtain the process requirements information of the target gravure printing plate steel plate, perform multi-dimensional data analysis and parameter derivation on the process requirements information, and obtain the preliminary process control parameter set of the target gravure printing plate steel plate. Historical process data matching the feature parameters of the target gravure printing plate steel plate is obtained. Based on the historical process data, time-series feature extraction and multi-dimensional data fusion are performed to extract sensitive key control variables in the steel plate rolling process. Based on the sensitive key control variables, the preliminary process control parameter set is optimized to obtain the rolling process parameter set. The rolling machine is controlled to perform rolling operation on the target gravure printing plate steel plate based on the rolling process parameter set. Real-time morphological data of the target gravure printing plate steel plate during the rolling process is obtained. The current process is identified based on the real-time morphological data, and error analysis and deviation extraction are performed between the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to obtain the morphological deviation data and process execution error information of the target gravure printing plate steel plate during the rolling process. If the shape deviation data exceeds the preset accuracy standard corresponding to the current process, adjust the rolling process parameter set based on the shape deviation data and process execution error information. Obtain the force and displacement response data of the target gravure printing plate steel plate, invert the material state data of the target gravure printing plate steel plate based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set. The process control target data is obtained, and multi-objective optimization calculations are performed based on the process control target data and the corrected rolling process parameter set to generate the target process control strategy. The rolling machine is then adjusted based on the target process control strategy.

[0008] In a third aspect, an electronic device is provided, comprising a processor, a memory, and a program or instructions stored in the memory and executable on the processor, the program or instructions being loaded and run by the processor to perform the steps of the above-described fully automated production method for rolling steel tubes for gravure printing plates.

[0009] In a fourth aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored therein, the program codes being adapted to be loaded and run by a processor to perform the steps of the above-described fully automated production method for rolling steel tubes for gravure printing plates.

[0010] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects: In implementing the technical solution of this invention, multi-dimensional data analysis and real-time feedback are used to optimize the rolling process parameters. By combining historical process data and real-time morphological data, the process parameters are dynamically adjusted to ensure the quality accuracy and production stability of the steel pipe, and to improve the adaptability and efficiency of the rolling process. Attached Figure Description

[0011] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Furthermore, similar numbers in the drawings are used to denote similar components, wherein: Figure 1 This is a schematic diagram of the first main structure of a fully automated production system for rolling steel tubes for gravure printing plates according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the second main structure of a fully automated production system for rolling steel tubes for gravure printing plates according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the main steps of a fully automated production method for rolling steel tubes for gravure printing plates according to an embodiment of the present invention. Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0012] Some embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0013] In the description of this invention, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and may also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.

[0014] See appendix Figure 1 , Figure 1 This is a first main structural block diagram of a fully automated production system for rolling steel tubes for gravure printing plates according to an embodiment of the present invention. Figure 1 As shown, a fully automated production system for rolling steel tubes for gravure printing plates, according to an embodiment of the present invention, mainly includes: The process requirements analysis module 101 is used to obtain the process requirements information of the target gravure printing plate steel plate, perform multi-dimensional data analysis and parameter derivation on the process requirements information, and obtain the preliminary process control parameter set of the target gravure printing plate steel plate. The process parameter determination module 102 is used to acquire historical process data that matches the feature parameters of the target gravure printing plate steel plate, extract time-series features and fuse multi-dimensional data based on the historical process data, extract sensitive key control variables in the steel plate rolling process, and optimize the preliminary process control parameter set based on the sensitive key control variables to obtain the rolling process parameter set. The deviation determination module 103 is used to control the rolling machine to perform rolling operation on the target gravure printing plate steel sheet based on the rolling process parameter set, obtain the real-time shape data of the target gravure printing plate steel sheet during the rolling process, identify the current process based on the real-time shape data, and perform error analysis and deviation extraction on the real-time shape data and the theoretical forming trajectory reference data corresponding to the current process to obtain the shape deviation data and process execution error information of the target gravure printing plate steel sheet during the rolling process. The adjustment module 104 is used to adjust the rolling process parameter set based on the shape deviation data and process execution error information if the shape deviation data exceeds the preset accuracy standard corresponding to the current process. The process parameter correction module 105 is used to acquire the force and displacement response data of the target gravure printing plate steel plate, invert the material state data of the target gravure printing plate steel plate based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set. The rolling adjustment module 106 is used to acquire process control target data, perform multi-objective optimization calculations based on the process control target data and the corrected rolling process parameter set, generate a target process control strategy, and adjust the rolling machine based on the target process control strategy.

[0015] In this embodiment, the target gravure printing plate steel plate is a raw material steel plate used to produce gravure printing plate steel pipes. This steel plate is a basic material that has undergone pretreatment such as cutting and forming and is ready for rolling.

[0016] Process requirements information is a collection of standards, parameters, and process requirements that the target gravure printing plate steel sheet must meet during the steel sheet rolling process. These requirements include the steel sheet's material, specifications, shape, dimensional accuracy, and other requirements, as well as the operating conditions and objectives during the rolling process, such as curvature, thickness, and roundness, to ensure the quality and consistency of the target steel sheet during the rolling process.

[0017] The preliminary process control parameter set is a set of control parameters initially set based on the process requirements of the target gravure printing plate steel sheet, as well as the mechanical properties and specifications of the relevant steel. These parameters include the operating settings of the rolling mill, such as pressure, temperature, and feed speed, and are designed to ensure that the target steel sheet can meet the expected forming effect and specification requirements during the rolling process.

[0018] During the acquisition and initialization of process requirements information, natural language processing and optical character recognition technologies are used to scan and parse design documents, process standards, and purchase orders, automatically extracting unstructured textual descriptive parameters. Simultaneously, CAD feature recognition technology is used to directly read annotation information and geometric features from 3D digital drawings. These extracted discrete data are then formatted and validated using feature mapping mechanisms and data cleaning techniques, removing outliers before being structured and stored in a database. This process extracts and forms steel plate material property data, including mechanical indicators such as yield strength and hardness, as well as steel plate specification data, including physical dimensions such as length, width, and thickness. These two sets of data are combined to obtain the preliminary process requirements dataset.

[0019] Based on the material properties of steel plates, the stress-response curves of steel plates are reconstructed in a simulation environment using digital twin virtual material testing technology. By performing spline interpolation and numerical gradient calculation on discrete stress-strain data points, the tangent modulus and real-time hardening rate of the material at different stress stages are dynamically obtained, thereby extracting plastic deformation resistance data that reflects the material's ability to resist plastic flow. Simultaneously, energy dissipation numerical integration technology is applied, combined with the material's elastic modulus and cross-sectional geometric parameters, to accurately calculate the total amount of elastic deformation energy released and the rebound direction within the material along the unloading path, thereby determining the springback characteristic parameters characterizing the degree of springback in cold working of the steel plate.

[0020] When pre-setting process accuracy based on steel plate specification data, 3D tolerance stack analysis technology is used to directly map the global tolerance boundaries in the target digital drawing to the 3D coordinate system of the rolling process. A discrete feature mapping matrix is ​​applied to decompose the global dimensional requirements according to spatial topological relationships, clearly defining the allowable control thresholds for specific dimensions such as roundness deviation, straightness, and wall thickness reduction rate, generating an accuracy requirement dataset. Based on this, spatial interpolation and motion trajectory smoothing techniques are used to transform these static dimensional specifications and shape preset parameters into dynamic envelopes that continuously change with processing time and spatial nodes, thereby generating theoretical forming trajectory benchmark data for subsequent comparison of the current process form and extraction of execution errors.

[0021] After integrating preset equipment constraint data, accuracy requirement datasets, plastic deformation resistance data, and springback characteristic parameters, a multi-level deformation gradient allocation strategy is used to decompose the overall forming task into multiple progressive processing passes to prevent excessive deformation during a single pressure application from causing material fracture. A case-based reasoning technique and feature similarity matching mechanism based on historical process big data are employed to retrieve the most recent real springback compensation history records from the production database that are closest to the current working condition, extracting the corresponding bending angle and roller pressing increment for springback compensation correction. Finally, a multi-objective heuristic optimization algorithm is used to globally coordinate and integrate control commands such as feed depth, pressing speed, and servo torque for each pass within the intersection space of the equipment's maximum load limit and the target accuracy requirement, ultimately outputting a preliminary set of process control parameters suitable for the target gravure printing plate steel sheet.

[0022] Characteristic parameters refer to physical or mechanical property data that describe the key properties and behaviors of the target gravure printing plate steel sheet. During the steel sheet rolling process, characteristic parameters include the material properties of the steel sheet, such as yield strength, elastic modulus, hardness, and plasticity, which reflect the deformation and load-bearing capacity of the steel sheet under external forces; geometric dimensions, such as the thickness, width, length, and curvature of the steel sheet, which determine the shape and processing difficulty of the steel sheet; and surface conditions, such as surface roughness and smoothness.

[0023] Historical process data is a dataset gradually accumulated over the past production process, detailing the entire processing flow and various related variables. This includes temperature, pressure, speed, mold shape, processing time, material properties, and measurement data from each stage of production.

[0024] Sensitive critical control variables are those that significantly impact the final product quality and production stability in the steel sheet rolling process. These mainly include parameters that need to be controlled during processing, such as temperature, pressure, speed, torque, stress, and strain. In the rolling process, sensitive critical control variables also involve the steel sheet's own material properties, such as its elastic modulus and yield strength; the equipment's operating conditions, such as roller pressure and speed; and changes in the production environment, such as temperature and humidity.

[0025] The rolling process parameter set is a collection of parameters that need to be controlled and adjusted during the rolling process of steel plates. It mainly involves aspects related to steel plate forming technology, equipment settings, and material properties. Specific parameters include the forming angle of the steel plate, roller pressure, processing speed, feed rate, forming temperature, and die geometry.

[0026] To obtain historical process data matching the characteristic parameters of the target gravure printing plate steel, the first step is to acquire the characteristic parameters of the target gravure printing plate steel. Specifically, relevant information is extracted from the steel plate's technical documents, design drawings, or real-time sensor data. These characteristic parameters include the steel plate's material properties, such as yield strength, elastic modulus, hardness, and plasticity, as well as geometric dimensions, such as thickness, width, length, and curvature. This data can be obtained through online sensor monitoring systems, product design software, or physical testing. Based on these characteristic parameters, similar historical production records are retrieved from the process database to obtain a correlated historical process dataset. Through the data interface of the distributed control system, SQL query commands with multi-condition join logic are used to extract raw records, including processing time, roller pressure, and mold geometry, from the real-time database and manufacturing execution system. During this process, box plotting is applied to identify and remove outliers caused by instantaneous sensor jitter, while linear interpolation is used to fill in sampling gaps caused by communication interruptions, ensuring data continuity in the time dimension. Finally, the cleaned data is normalized to map material properties and processing variables of different physical dimensions to the 0 to 1 range, thus constructing a structured historical process data base.

[0027] Based on cleaned historical data, a sliding window algorithm was used to segment the temperature, torque, and strain sequences along the time axis. Fast Fourier Transform was employed to extract the dominant frequency components during the processing, thereby identifying the periodic fluctuations caused by the roller rotation speed. Simultaneously, higher-order statistical moment analysis was used to calculate the mean, variance, and kurtosis within each time window, quantifying the stress response characteristics of the material in the initial stage of rolling. Subsequently, normative correlation analysis was used to fuse the extracted temporal features with spatial measurement data in a multidimensional manner. A vector space mapping mechanism was used to transform discrete sensor feedback into a fused feature vector with spatiotemporal coupling properties, providing full-dimensional data support for identifying the dynamic evolution logic during the rolling process.

[0028] When extracting the key control variables sensitive to steel plate rolling, a feature importance analysis technique based on gradient boosting decision trees is used, integrating feature vectors as input. By calculating the information gain contribution of each variable in the decision branches, the core factors significantly affecting the roundness and dimensional accuracy of the final steel pipe are identified. Principal component analysis is used to reduce the dimensionality of the high-dimensional variable space, eliminating redundant interference terms, and focusing on extracting key control variables such as elastic modulus, roller pressure, yield strength, and real-time torque. Simultaneously, Spearman's rank correlation coefficient is used to quantitatively assess the nonlinear coupling strength between variables, ultimately selecting a set of variables reflecting the essential characteristics of the rolling process, achieving precise identification of the core driving factors in steel pipe forming. The feature importance analysis technique based on extreme gradient boosting trees is primarily used to train an extreme gradient boosting tree regression model to predict key process results in steel plate rolling, such as roundness and dimensional accuracy. After model training, the contribution of each input feature to the prediction result is calculated.

[0029] When optimizing the preliminary process control parameter set based on extracted sensitive key control variables, a constrained particle swarm optimization algorithm is used to perform a heuristic search in the multidimensional solution space. This process uses material hardness and strain hardening index from the sensitive key control variables as nonlinear compensation terms, employs finite element simulation regression technology to predict the springback amount after steel plate unloading, and dynamically corrects the initial design forming angle and roller pressure settings. For the coordinated balance between processing speed and feed rate, the Actor-Critic framework in deep reinforcement learning is used, mapping production efficiency and residual stress as feedback signals to the reward function. Through a policy network, policy gradient ascent and closed-loop iterative calculations are performed in the dynamic environment, ultimately encapsulating the optimized values ​​into a rolling process parameter set. The reward function is a key component in deep reinforcement learning used to measure the effectiveness of the policy. In the rolling process optimization, the reward function is calculated based on key indicators such as production efficiency and residual stress, and these indicators are input into the system as feedback signals. By evaluating the execution effect of the current process, the reward function provides positive or negative feedback to the policy network, thereby driving policy adjustment and optimization. Ultimately, the reward function guides the reinforcement learning model to find the optimal process parameters to ensure that the accuracy and efficiency of the steel plate rolling process are improved.

[0030] A rolling machine is a piece of equipment used for forming and processing steel pipes. Its function is to roll steel plates into circles or specific curved shapes according to set radius and angle requirements.

[0031] The current process refers to the specific processing stage or step in the rolling process of a steel sheet. Each process may include different operations, such as roller pressing, heating, cooling, and shape adjustment, and each process has its unique operational requirements and objectives. In the rolling process, the current process refers to the step that is being performed or has been completed, which determines the current forming state of the steel sheet.

[0032] Real-time morphological data refers to the geometric morphology information of the steel plate monitored and recorded in real time during the steel plate rolling process using various sensors and imaging technologies. It can intuitively reflect the key features such as the real-time shape, size, angle, and curvature of the steel plate during the rolling process.

[0033] Theoretical forming trajectory reference data refers to the theoretical data for the ideal or preset path, shape, and dimensions of the steel plate forming process in each processing step. Each step has different target requirements and operations during the rolling process, therefore each step will have specific reference data.

[0034] Shape deviation data refers to the discrepancy between the actual shape of the steel sheet during the rolling process and the theoretical forming trajectory reference data corresponding to the current process. It mainly includes shape errors, such as the degree of deviation from a circle, and dimensional errors, including deviations in the steel sheet's diameter, bending angle, curvature, etc. In addition, it also includes deviations in other geometric parameters, such as the uniformity of the steel sheet wall thickness and the flatness of the curved surface.

[0035] Process execution error information refers to the deviation between the actual process execution and the preset or ideal process parameters during the steel plate rolling process. Specifically, it reflects the difference between the actual operation and the target forming trajectory reference data in each process, including errors in equipment accuracy, pressure, speed, material deformation, etc.

[0036] After presetting the rolling shape and dimensional accuracy based on steel plate specifications and obtaining the accuracy requirement dataset, multi-axis kinematic trajectory generation technology is used to transform the static three-dimensional geometric tolerance boundaries into a spatial coordinate flow that dynamically changes with the processing sequence. By applying spatial curve interpolation technology, the discrete tolerance requirements of each control node are smoothly connected in three dimensions, drawing an ideal curvature gradient path and cross-sectional evolution profile for each specific processing pass. These continuous geometric paths with time labels and spatial physical constraints constitute the theoretical forming trajectory reference data used to guide and verify each subsequent processing step.

[0037] During the rolling operation, the modified rolling process parameters are converted into drive instructions that can be directly executed by the programmable logic controller (PLC) using low-level communication protocol parsing technology, precisely controlling the servo spindle speed and the hydraulic cylinder's pressing depth. During the bending process of the steel plate, high-frequency laser triangulation or structured light 3D scanning technology is applied to perform non-contact continuous surface reconstruction of the forming area, densely acquiring discrete point arrays of 3D spatial coordinates of the inner and outer surfaces of the tube wall. This yields real-time morphological data that intuitively reflects the current bending angle, instantaneous curvature, and surface physical undulations.

[0038] After acquiring the real-time 3D coordinate matrix, geometric feature contour extraction technology is used to calculate the global radius of curvature, end warpage, and cross-sectional closure gap of the current steel plate. These real-time extracted physical feature vectors are dynamically compared with a pre-stored multi-pass processing sequence list. A dynamic time warping algorithm is applied to calculate the topological similarity between the current real-time morphological features and the benchmark sequences of each standard process. By locking the processing interval where the similarity peak is located, the current stage of the steel plate—whether it is in the initial end pre-bending, multi-pass main rolling, or final seam pressing and rounding stage—is accurately identified, thus clarifying the current process.

[0039] After identifying the current process, the theoretical forming trajectory reference data that strictly corresponds to this process is extracted. Using iterative nearest-point spatial registration technology, the 3D morphological point cloud obtained from real-time scanning is rigidly aligned with the theoretical reference surface under a unified global coordinate system. After alignment, 3D normal distance calculation technology is applied to calculate the true vector distance of the measured physical surface from the ideal reference surface point by point along the normal direction of the theoretical surface. This precise set of numerical values ​​quantifying the overall ellipticity deviation, local surface unevenness, and bending angle springback allowance is extracted as the morphological deviation data of the target gravure printing plate steel sheet under the current process.

[0040] Based on the extracted morphological deviation data, kinematic inverse analysis technology is used to map physical distortions in geometric space back to the mechanical motion execution domain of the rolling machine. By applying spatiotemporal error backpropagation tracing, specific geometric defects, such as circumferential ripples or axial deflection, are strictly aligned with the historical action logs at the equipment's underlying level in terms of timestamp dimension and causal tracing. This allows for precise identification of whether the specific shape deviation is caused by hydraulic cylinder pressure lag, asynchronous feed speeds of the two spindles, or fluctuations in servo torque output. Ultimately, this data, revealing the dynamic disconnect between the actual mechanical motion response of the equipment and the preset control commands, is extracted and summarized as process execution error information.

[0041] Preset accuracy standards are the target shape and dimensional accuracy requirements set for each step in the rolling process. They are derived through theoretical calculations, design requirements, or historical data analysis, and are used to guide control indicators during processing. Preset accuracy standards define the ideal processing results for each step, such as the roundness, diameter, curvature, and bending angle of the steel plate.

[0042] When the morphological deviation data exceeds the preset accuracy standard corresponding to the current process, the out-of-tolerance quantities in dimensions such as diameter, roundness, and angle are first classified using a fuzzy logic judgment algorithm, while simultaneously retrieving process execution error information. By applying cross-correlation analysis, the correlation weights between each geometric deviation component and pressure fluctuations and speed lag at the equipment execution end are calculated, thereby identifying the main causes of the deviation. Based on this, a mapping relationship between morphological deviation and sensitive key control variables is established using a Jacobi sensitivity matrix, quantitatively separating the fixed deviation caused by mechanical clearance from the random deviation caused by material property fluctuations, providing precise logical input for parameter compensation.

[0043] To identify the root cause of the deviation, the system uses the modified Newton's method to perform a reverse iterative solution in the parameter space, calculating the pressure increment and forming angle compensation value required to offset the morphological deviation. During this process, a recursive least squares method is introduced to estimate the online springback coefficient of the material in real time, which is then used as a dynamic disturbance term in the decoupled control equations to correct the nonlinear compensation error caused by hardness fluctuations. Through this closed-loop optimization iteration based on error feedback, the system transforms static geometric deviations into parameter adjustment vectors with physical meaning, ensuring that the compensation logic can accurately adapt to the current processing conditions.

[0044] Finally, the calculated adjustment vectors are superimposed on the original parameters, and weighted feedforward compensation technology is used to update the roller pressure, feed speed, and forming angle in the rolling process parameter set in real time. To prevent system oscillations caused by large adjustments, a first-order hysteresis filter operator is applied to smooth the updated data stream, and a quadratic boundary check based on the Kuhn-Tak condition is performed before output to ensure that the adjusted parameters are strictly within the safe operating envelope of the equipment. Through this dynamic correction mechanism driven by morphological deviations exceeding the preset accuracy standard corresponding to the current process, the system generates an optimized rolling process parameter set.

[0045] Force and displacement response data are the mechanical response data of the steel plate after being subjected to force, collected in real time by sensors or measuring equipment during the rolling process. Force response data mainly includes various external forces acting on the steel plate, such as pressure, tension, and torque, while displacement response data records the deformation of the steel plate under these external forces, such as displacement, bending, and tension.

[0046] Material condition data is derived from the inversion analysis of force and displacement response data obtained during the rolling process of steel plates, thereby inferring the internal material properties and mechanical state of the steel plates. It mainly involves characteristics such as the elastic-plastic state, hardness, yield strength, and stress-strain relationship of the steel plates, reflecting their deformation capacity and resistance during the forming process.

[0047] To acquire force and displacement response data of the target gravure printing plate steel sheet, high-precision tensile and compressive sensors and torque sensors deployed on the hydraulic actuator and spindle of the rolling machine are first used to continuously collect dynamic contact loads and driving torques during the forming process. Simultaneously, a non-contact laser displacement sensor and a high-frequency optical grating ruler are used to track the bending deformation trajectory and relative displacement in three-dimensional space of the steel sheet surface in real time. The acquired raw physical signals are transmitted via a high-speed data acquisition channel, and Kalman filtering technology is applied to filter out high-frequency dynamic noise introduced by equipment mechanical vibration and electromagnetic interference. Subsequently, a high-precision clock synchronization mechanism is used to strictly align the load sequence and deformation sequence on the time axis, generating force and displacement response data that accurately maps the coupling relationship between transient force and deformation of the steel sheet.

[0048] After acquiring the force and displacement response data, it is imported into a low-level computational framework based on finite element inversion to deeply analyze the hidden mechanical state inside the steel plate. This stage utilizes the Levenberg-Marquardt nonlinear least squares optimization algorithm to iteratively adjust the initial variables in the preset material constitutive equations until the stress-deformation curves calculated by the system simulation highly coincide with the measured force and displacement response data, and the set convergence accuracy is achieved. Through this inverse parameter solution under boundary constraints, the system directly extracts and isolates the optimally fitted real-time yield strength, elastic modulus attenuation rate, and true stress-strain curves, encapsulating these quantitative indicators into material state data reflecting the current plastic deformation capacity and resistance of the pipe.

[0049] When generating the final correction command, the extracted material state data and the adjusted rolling process parameter set are concatenated using feature tensors. Gaussian process regression technology is used to construct a correlation mapping between the nonlinear mechanical characteristics of the material and the final forming springback. By performing partial correlation calculations on this mapping, the perturbation weights of real-time yield strength fluctuations or hardness dispersion on the current forming angle and roller pressure are quantitatively removed. Based on the predicted curvature output of the regression analysis, a dynamic feedforward compensation mechanism is used to deduce the additional downward pressure increment and over-bending angle compensation values ​​required to accurately overcome the current material differences. These compensation vectors are directly superimposed and injected into the adjusted process parameter set to generate a corrected rolling process parameter set that can adapt to local material property fluctuations.

[0050] Process control target data refers to the target data collected by various sensors and monitoring systems during the steel plate rolling process, used to evaluate the quality of the rolling process. This data includes key performance indicators such as the roundness, surface roughness, dimensional accuracy, and wall thickness uniformity of the steel pipe, corresponding to the target requirements of the rolling process.

[0051] A target process control strategy is a solution that optimizes process control objectives. This strategy involves systematically adjusting the operating parameters of the rolling mill, including roller pressure, speed, and pressure angle, to achieve an optimal balance among various process objectives. The target process control strategy not only clarifies the adjustment methods for current process parameters but also develops solutions to address various potential disturbances and adapt to fluctuations in the properties of different materials, ensuring that the precision and quality of the final product meet standards.

[0052] When acquiring process control target data, a high-precision 3D coordinate measuring machine and an online laser profilometer deployed at the end of the production line are first used to perform a full circumferential scan of the formed steel pipe to obtain an original point cloud that reflects its geometric shape. The roundness deviation of the steel pipe is extracted from the point cloud using the minimum area circle method, and white light interferometry is used to capture the surface micro-morphology to calculate the average roughness value. Simultaneously, an ultrasonic thickness gauge is used for multi-point sampling, and the uniformity distribution of the wall thickness is evaluated through a spatial interpolation algorithm. Finally, these measured values, including key indicators such as roundness, roughness, and wall thickness, are structured and integrated to form process control target data that can quantify product quality requirements.

[0053] Based on the process control target data and the corrected rolling process parameter set, a Pareto optimal solution search is performed within a multi-dimensional decision space consisting of roller pressure, rotational speed, and pressurization angle using a non-dominated sorting genetic algorithm with crowding distance. In this process, the analytic hierarchy process (AHP) is introduced to assign dynamic weights to roundness accuracy and production efficiency, while a penalty function technique is used to transform the nonlinear fluctuations of the material state into soft constraints on the search boundary. Monte Carlo sensitivity verification is performed in a digital twin model to simulate the parameter disturbance resistance under different operating conditions, thereby locking in the optimal balance point among multiple competing objectives and ultimately generating a target process control strategy that includes parameter adjustment benchmarks and disturbance compensation mechanisms.

[0054] During adjustment, the system utilizes the Industrial Ethernet protocol to map the instruction set in the target process control strategy to the register address data of the rolling machine PLC in real time. Applying feedforward-feedback composite control technology, the roller pressure setpoint in the strategy is used as a preset gain, synchronously combined with the real-time load current fed back from sensors, and a fuzzy adaptive PID algorithm is used to finely adjust the proportional relief valve of the hydraulic pump station. During this period, dynamic instruction flow smoothing technology controls the acceleration and deceleration slope of the motor speed to ensure seamless connection between roller pressure switching and pressure angle changes. Through this strategy-driven physical execution, the rolling machine can automatically complete closed-loop correction of mechanism posture and power output according to real-time operating conditions, ensuring that the produced gravure printing plate steel pipes always approach the preset quality target under dynamic conditions.

[0055] In this embodiment, the rolling process parameters are optimized through multi-dimensional data analysis and real-time feedback. By combining historical process data and real-time morphological data, the process parameters are dynamically adjusted to ensure the quality accuracy and production stability of the steel pipe, and to improve the adaptability and efficiency of the rolling process.

[0056] Based on the above technical solution, optionally, the process requirements analysis module 101 is specifically used for: Multidimensional data parsing and parameter initialization are performed on the process requirement information to obtain a preliminary process requirement dataset; the preliminary process requirement dataset includes steel plate material property data and steel plate specification data; Based on the material property data of the steel plate, mechanical property analysis was performed to derive the plastic deformation resistance data and springback characteristic parameters of the target gravure printing plate steel plate. Based on the steel plate specification data, the shape and size accuracy during the rolling process are preset to obtain the accuracy requirement dataset; Based on preset equipment constraint data, accuracy requirement dataset, plastic deformation resistance data, and springback characteristic parameters, multi-pass deformation amount allocation and parameter integration are performed, and springback compensation correction is performed based on preset springback prediction model to obtain the preliminary process control parameter set of the target gravure printing plate steel plate.

[0057] In this scheme, the preliminary process requirements dataset is a basic dataset formed after preliminary analysis of the steel plate material and specifications before the rolling process begins. This includes the steel plate's material properties, such as strength, hardness, and modulus of elasticity, as well as its dimensional parameters, such as diameter, wall thickness, and length.

[0058] Steel plate material property data describes the material characteristics of steel plates, including mechanical property parameters such as yield strength, elastic modulus, hardness, and plasticity. These parameters can reflect the deformation, load-bearing capacity, and processing performance of steel plates under external forces.

[0059] Steel plate specifications are data related to the geometric dimensions of steel plates, including physical parameters such as outer diameter, inner diameter, wall thickness, and length.

[0060] Plastic deformation resistance data refers to the ability of a steel plate to resist plastic deformation during the rolling process, and is usually determined by the material's inherent mechanical properties such as yield strength and hardness. The magnitude of the force required during rolling and the deformation process of the steel plate are also affected.

[0061] Springback characteristics refer to the springback behavior of steel sheets during cold working processes such as rolling, which results in the release of internal stress. Common springback characteristics include springback angle and springback rate.

[0062] The accuracy requirement dataset consists of the morphological and dimensional accuracy standards that need to be achieved during the steel plate rolling process. This includes the geometric accuracy of the steel plate's shape, such as roundness, diameter, and wall thickness consistency, as well as dimensional accuracy, such as straightness and flatness.

[0063] Preset equipment constraint data refers to the technical specifications and operational limitations of processing equipment such as rolling machines, including the equipment's maximum load capacity, working stroke, minimum adjustment accuracy, and operating speed. These data limit the range of process parameter settings and the actual execution effect.

[0064] The pre-defined springback prediction model is a mathematical model for predicting springback based on known data of the steel plate's material and geometric parameters. This model can predict the amount of springback during the rolling process, taking into account the steel plate's material, specifications, and processing method.

[0065] When performing multi-dimensional data analysis and parameter initialization of process requirements information, natural language processing technology is used to extract text features from unstructured process documents, and CAD drawing parsing algorithms are used to automatically identify the cross-sectional geometric parameters of the steel plates. During this process, the extracted characters and values ​​are standardized using a feature mapping mechanism, and data cleaning techniques are employed to remove outliers or incomplete annotations. Subsequently, the validated parameters are structured and stored, and mechanical constants from the material property database are retrieved synchronously to dynamically generate steel plate material property data, including mechanical indicators such as steel plate strength, hardness, and elastic modulus, as well as steel plate specification data, including physical dimensions such as outer diameter, wall thickness, and length. These together constitute the preliminary process requirements dataset.

[0066] Based on the material properties of steel plates, spline interpolation technology is used to reconstruct continuous surfaces from the measured discrete data points of uniaxial tensile stress and strain, directly capturing the true mechanical characteristics of the material. By performing numerical gradient calculations at different strain nodes, the tangent modulus and real-time hardening rate of the material are dynamically obtained, thereby deriving the strength evolution path reflecting the true plastic rheological behavior of the material, and thus quantifying the plastic deformation resistance data during steel plate processing. Simultaneously, energy dissipation numerical integration technology is applied, combined with the material's elastic modulus and moment of inertia of the cross section, to accurately calculate the total amount of elastic strain energy released inside the material through unloading path integration, thereby determining the springback characteristic parameters characterizing the degree of springback in cold working of the steel plate.

[0067] When setting precision based on steel plate specification data, CAD model feature recognition technology is used to extract the geometric dimensions and tolerance design indicators embedded in the 3D CAD model of the target gravure printing plate steel plate, serving as the original basis for defining precision boundaries. Through tolerance stack analysis technology, the theoretical tolerance boundaries at the design end are directly mapped to the machining coordinate system, automatically matching the geometric tolerance level of the target steel plate. Using a discrete feature mapping matrix, global dimensional requirements are directly mapped to specific technical indicators such as roundness, straightness, and wall thickness consistency through logical indexing. With the help of preset correspondences, a linear decomposition from specifications to local features is achieved. In this stage, error propagation analysis is used to simulate the influence weight of machining tolerances on the final shape under different passes, and the allowable deviation range for each physical dimension is set through sensitivity matrix calculation. The entire process transforms static steel plate specifications into dynamic process control thresholds, effectively eliminating the logical gap between design and manufacturing. Finally, the calculated indicators are logically encapsulated to generate a precision requirement dataset including geometric precision standards and dimensional precision standards.

[0068] In generating the preliminary process control parameter set, firstly, based on preset equipment constraint data, a multi-level logarithmic strain distribution rule is applied to scientifically allocate the total deformation to multiple processing passes. A preset springback prediction model is used to calculate the instantaneous shape after each pass in real time, and a radius difference compensation method is employed to correct the roller feed depth for springback. Through a simplex optimization algorithm, parameter integration is performed within the intersection space of equipment power limits and accuracy requirements to automatically determine the optimal roller pressure, feeding speed, and support position. Finally, these corrected and verified instructions are logically encapsulated to obtain the preliminary process control parameter set for the target gravure printing plate.

[0069] The training process for the preset rebound prediction model is as follows: Historical processing data and relevant material data of the target gravure printing plate steel sheet were collected, including the steel sheet's mechanical properties and geometric characteristics. Mechanical properties included yield strength, elastic modulus, and hardness, while geometric characteristics involved parameters such as outer diameter, wall thickness, and length. This data served as the input features for the model, forming the foundation for the springback prediction model. Based on the steel sheet's mechanical and geometric data, the finite element method was used to simulate the potential internal stress of the steel sheet during different processing stages. Numerical simulation was employed to calculate the stress distribution during the cold working process, focusing on stress release and springback effects, while simultaneously recording the springback amount under different processing conditions. The simulation results were used as training data for model optimization and adjustment. To improve springback prediction accuracy, regression analysis and machine learning algorithms were introduced to learn from and fit the historical data. Machine learning algorithms included support vector machines and random forests. By comparing simulated springback data with actual springback data, the algorithm's internal weight parameters were optimized until the prediction error was minimized. In this process, backpropagation and cross-validation techniques are used to verify the model's generalization ability, ensuring that the model can adapt to springback scenarios of steel plates with different materials, specifications, and processing conditions. Through the above series of processes, a springback prediction model is finally constructed. This model can accurately predict the springback amount during the rolling process of steel plates based on real-time material and specification data.

[0070] In this solution, by comprehensively considering the steel plate material, specifications, and historical data, the rolling process parameters are optimized, the springback effect is effectively derived and compensated, and the forming quality and dimensional accuracy of the steel pipe are ensured.

[0071] Based on the above technical solution, optionally, the process parameter determination module 102 is specifically used for: Extract dynamic evolution features from historical process data to form a historical time-series feature dataset of the steel plate rolling process; Feature engineering and information fusion are performed on historical time-series feature datasets to obtain a comprehensive historical dataset of the steel plate rolling process. High-dimensional feature importance assessment was performed based on historical comprehensive datasets to extract sensitive key control variables in the steel plate rolling process; Based on the sensitive key control variables, a multivariate coupled correlation analysis was conducted to obtain the set of factors affecting the rolling quality; Based on the set of factors affecting the quality of rolling, the preliminary process control parameter set is optimized and adjusted in multiple objectives to obtain the rolling process parameter set.

[0072] In this scheme, dynamic evolution characteristics refer to the key process features that continuously change over time during the steel plate rolling process. These include time-series data related to the mechanical response, deformation process, and temperature changes during rolling.

[0073] Historical time-series feature datasets are collections of feature data with time-series characteristics extracted after analyzing historical data of the steel plate rolling process. They encompass all process parameters and measurements arranged in chronological order, such as temperature, pressure, and deformation, comprehensively recording the changes in process parameters at different points in time and the specific impacts of these changes on the production process.

[0074] Historical comprehensive datasets are datasets formed by integrating historical time-series data from multiple sources using feature engineering and information fusion techniques. They aggregate different types of data, including process parameters, sensor data, and equipment status information, providing a more comprehensive view of various influencing factors in the rolling process.

[0075] The set of factors affecting the quality of steel pipe rolling is a collection formed after sorting out all relevant factors that affect the forming quality of steel pipe during the rolling process. These include various factors such as steel plate material, rolling machine operating parameters, operating conditions, and external environment. Through systematic analysis and identification, the combination of factors that plays a key role in the final characteristics of steel pipe, such as roundness, surface quality, and dimensional accuracy, is finally selected.

[0076] When extracting dynamic evolution features from historical process data, a sliding window sampling technique is used to continuously slice the raw pressure, temperature, and displacement data streams collected by various sensors to capture transient changes in mechanical response and thermodynamic state. Fast Fourier Transform is applied to extract periodic frequency components from spindle torque and equipment vibration, and a differential algorithm is used to calculate the instantaneous evolution slope of deformation rate and temperature gradient. The extracted time-varying features, such as peak values, root mean square (RMS), and dynamic slopes, are rigorously synchronized and aligned using high-precision timestamps to generate a historical time-series feature dataset that is tightly arranged in time sequence and records the process response trajectory.

[0077] When performing feature engineering and information fusion on historical time-series feature datasets, a nonlinear interaction term between temperature, pressure, and deformation rate is constructed using polynomial cross-expansion technology to quantify the implicit correlations under the synergistic effects of multiple physical fields. Multidimensional adaptive filtering and spline interpolation techniques are employed to perform joint denoising and missing value filling on multi-source heterogeneous time-series signals, eliminating biases caused by differences in sensor sampling frequencies. Subsequently, tensor stitching technology is used to integrate the cleaned time-series feature matrix with discrete static equipment initial states and material batch attributes in a high-dimensional space, breaking down the physical dimensional barriers between heterogeneous data and constructing a comprehensive historical dataset that presents a panoramic view of potential influencing factors.

[0078] When evaluating the importance of high-dimensional features and extracting key variables from a historical comprehensive dataset, a node splitting contribution evaluation technique based on gradient boosting decision trees is used to directly quantify the physical contribution weights of each input node, such as elastic modulus, yield strength, and real-time torque, to the final roundness and dimensional deviations. A recursive feature elimination strategy and collinearity testing are applied to remove redundant environmental interference terms, focusing on retaining the core parameters that dominate geometric distortion and stress distribution during processing. This allows for the precise extraction of sensitive key control variables that determine the final product quality and production stability.

[0079] When performing multivariate coupling correlation analysis on sensitive key control variables, partial correlation matrix analysis is used to calculate the independent mapping strength between material property fluctuations, equipment roller pressure, and external environmental temperature and humidity, excluding interference from other variables. Topological network mapping technology is combined to characterize the nonlinear synergistic effects and feedback paths among multiple variables, and weak correlations are filtered out by setting a significance threshold. Combinations of variables exhibiting strong coupling and directly driving the deterioration of surface quality and dimensional accuracy are logically categorized, ultimately determining a set of factors influencing rolling quality, including core materials, equipment status, and operational dimensions.

[0080] When optimizing and adjusting the initial process control parameter set using a set of factors influencing rolling quality, a multi-objective evolutionary optimization algorithm is employed to perform Pareto front search in a multi-dimensional decision space encompassing production stability and molding accuracy. Extracted material characteristic fluctuations and environmental temperature and humidity are introduced as dynamic disturbance terms into the constraint boundary, and virtual simulation iteration technology is used to limit unrealistic aggressive adjustments. Dynamic feedforward compensation technology is applied to superimpose the calculated pressure fine-tuning value and speed correction coefficient into the initial parameters. After smoothing and filtering to eliminate abrupt commands, the final output is a rolling process parameter set that can adapt to various nonlinear fluctuations and ensures that molding quality meets standards.

[0081] In this solution, by extracting and analyzing historical process data, sensitive key control variables in the steel plate rolling process are identified, and rolling process parameters are optimized, thereby improving product quality and process accuracy.

[0082] Based on the above technical solution, optionally, the deviation determination module 103 is specifically used for: Based on the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process, spatial registration and difference calculation are performed to obtain the morphological deviation data during the rolling process of the target gravure printing plate steel plate. Error feature extraction and deviation analysis are performed based on morphological deviation data to obtain the spatial distribution pattern of morphological errors. Based on the spatial distribution pattern and the preset error propagation mechanism, source tracing analysis and evolutionary deduction are performed to obtain process execution error information during the rolling process of the target gravure printing plate steel plate.

[0083] In this scheme, the spatial distribution pattern describes the distribution law of morphological errors on the surface or cross-section of the steel plate, clarifying the intensity and direction of the errors at different locations. By conducting spatial analysis on the morphological deviation data, the distribution characteristics of the errors on the entire surface of the steel plate or in specific areas can be determined, such as concentrated areas or uniform distribution states of the errors.

[0084] A pre-defined error propagation mechanism is a mechanism that predicts and describes the propagation or amplification of errors at different processing stages by combining the physical laws, mechanical models, or historical process experience in the steel plate rolling process. For example, it describes how factors such as springback, plastic deformation, and compressive stress affect the geometry of the steel plate, and how these errors are gradually transmitted from the initial processing stage to the final product shape.

[0085] Based on the acquired real-time morphological data, the iterative nearest-point algorithm is used to perform fine registration between the discrete point cloud acquired by 3D scanning and the theoretical forming trajectory reference data corresponding to the current process in a unified global coordinate system, achieving physical alignment between the measured contour and the design reference. After registration, spatial point-to-surface distance calculation technology is used to calculate the geometric deviation between the measured point matrix and the ideal trajectory point by point along the normal vector direction of the theoretical surface. By structurally integrating these differences, including local concavity / convexity, overall roundness deviation, and curvature fluctuations, precise quantitative morphological deviation data of the current forming state from the target is generated.

[0086] After extracting morphological deviation data, spatial gradient field analysis is applied to calculate the evolution rate of deviations in the axial and circumferential directions of the steel plate, thereby clarifying the intensity and extension direction of errors at different surface locations. Simultaneously, a density-based clustering algorithm is introduced to regionalize the deviation values, automatically separating distorted regions with abnormally concentrated errors from uniformly distributed offset regions. By mapping these identified error features onto the surface topological mesh of the steel plate, the clustering patterns and geometric characteristics of errors with varying locations are depicted, thus forming a concrete spatial distribution pattern of morphological errors.

[0087] Based on the acquired spatial distribution pattern and combined with a pre-set error propagation mechanism, inverse physical mapping technology is used to project the geometric distribution features of the surface back into the operation instruction chain of the rolling process. By applying feature map comparison technology, the current distribution pattern is matched with the springback evolution and stress accumulation path recorded in the pre-set propagation mechanism to trace the physical evolution trajectory of the error from the initial stage of processing to the final shape. Using this time-series causal inference logic, the underlying action sources causing abnormal morphology are identified in depth, such as uneven roller pressing force, feed speed pulsation, or material property fluctuations. Finally, specific process execution error information, including mechanical response hysteresis and deviation of process parameters, is extracted.

[0088] In this solution, the morphological error of the steel plate is extracted and its spatial distribution is analyzed by morphological data registration and difference calculation. Combined with the error propagation mechanism, the source is traced and deduced to accurately identify the process execution error.

[0089] Based on the above technical solution, optionally, the process parameter correction module 105 is specifically used for: Based on the material state data, rheological features reflecting the material's rheological behavior and elastoplastic features reflecting the material's elastoplastic properties are extracted respectively. The rheological features, elastoplastic features, and the adjusted rolling process parameter set are subjected to cross-domain data alignment and modal fusion to construct a standardized dataset with enhanced features. A global sensitivity analysis was performed based on the standardized dataset to identify the contribution factors of rheological characteristics to the rolling execution deviation. Based on the contribution factor and the adjusted rolling process parameter set, a nonlinear coupling mechanism analysis was performed to obtain the synergistic effect data and interaction data between the material's elastic-plastic state and the rolling process parameters. Based on the synergistic effect data and interactive influence data, a process compensation proxy model is constructed. Based on the process compensation proxy model, incremental prediction and compensation analysis are performed on the adjusted rolling process parameter set to obtain the corrected rolling process parameter set.

[0090] In this scheme, rheological characteristics are parameters that reflect the deformation behavior of materials under stress or strain, including yield strength, flow stress, strain hardening rate, etc. These characteristics can reveal the mechanical response of materials in plastic or viscous flow states, helping to describe the deformation characteristics of materials during processing.

[0091] Elastic-plastic properties are parameters that describe the deformation behavior of a material in the elastic and plastic stages, including indicators such as elastic modulus, yield strength, and strain hardening coefficient. These properties reflect the material's ability to deform and recover under external forces, and are particularly evident in cold working processes such as rolling steel plates.

[0092] A standardized dataset is a dataset that has undergone preprocessing and standardization. By normalizing or standardizing features of different dimensions and scales, scale differences between various types of data are eliminated, allowing the data to be analyzed and trained within the same measurement system.

[0093] Contribution factors are the degree and weight of the influence of various variables on the target outcome in multivariate analysis. A typical target outcome is the rolling execution deviation.

[0094] Synergistic effect data refers to the combined effect data generated when multiple factors act on a system. In the rolling process, the synergistic effect between the material's elastic-plastic state and process parameters is reflected as the overall influence formed by their interaction.

[0095] Interactive effect data refers to the composite effect data caused by the interaction of various factors within a multivariable system. During the rolling process, the interaction between process parameters such as roller pressure and feed speed and material properties affects the final forming quality of the steel pipe.

[0096] The process compensation proxy model is a mathematical model that predicts processing deviations and provides compensation schemes based on process parameters and material properties. This model outputs an adjusted set of process parameters through prediction and compensation analysis of process deviations.

[0097] Based on material state data, the slope of the inverted real stress-strain curve is calculated point-by-point using local gradient numerical differentiation technology. The hardening slope and flow stress level of the plastic flow stage are extracted and solidified into rheological characteristics reflecting the material's rheological behavior. Simultaneously, the secant modulus evaluation method, combined with linear back-calculation data from the strain recovery zone, is used to determine the critical stress point and energy dissipation rate of the material's transition from elastic to plastic, thereby extracting elastoplastic characteristics including the elastic modulus and yield criterion. These characteristics characterize the flow law and rebound properties of the pipe under the current stress state, revealing its mechanical essence.

[0098] When integrating the acquired rheological and elastoplastic features with the adjusted rolling process parameter set, spatiotemporal indexing and association techniques are used to align the multi-source heterogeneous mechanical parameters with the equipment execution commands according to the processing step completion time benchmark. Z-score normalization is employed to center the hardness, pressure, and rotational speed values ​​of different dimensions, eliminating the bias effect caused by magnitude differences in subsequent calculations. Then, through multimodal feature concatenation, the cleaned mechanical matrix and parameter tensor are merged in high dimension. After feature denoising and smoothing to remove discrete noise, a feature-enhanced normalized dataset with strong expressive power is finally constructed.

[0099] Based on a standardized dataset, variance decomposition technology is used to calculate the percentage of output fluctuation explained by input variables in the global parameter space, thereby identifying the contribution factors of rheological characteristics to the rounding execution deviation. In this process, an elimination-based sensitivity test is implemented. By simulating the change trajectory of residual error under small perturbations of rheological parameters, the weighting ratio of strain hardening rate and flow stress on the final roundness deviation is quantitatively determined. This step abstracts the complex mechanical response into specific weight values, clarifying the role of material fluctuations in the execution deviation.

[0100] When conducting nonlinear coupling mechanism analysis based on contribution factors and process parameter sets, the partial effect surface extraction technique is used to capture the nonlinear response law under the interplay of material hardening state and roller pressure. Through the principle of vector superposition, the total energy dissipation distribution when elastoplastic restoring force and mechanical pressing force act together on the pipe is calculated, deriving the synergistic effect data formed by their combined action. Simultaneously, the main effect independent mapping method is introduced to eliminate the linear contribution of single variables, specifically extracting the nonlinear abrupt change characteristics caused by pressure changes constrained by material strength, i.e., the interactive influence data, thereby revealing the differences in the effectiveness of process regulation under different material states.

[0101] When constructing the process compensation surrogate model, radial basis function mapping technology is used to transform the synergistic effect and interaction data into a predictive numerical kernel. By establishing a nonlinear topological relationship between material fluctuations and parameter compensation amounts, a process compensation surrogate model capable of predicting processing deviations in real time is finally trained and generated. During model operation, residual error iterative correction technology is used to dynamically incrementally predict the adjusted rolling process parameter set, calculating the pressure fine-tuning value and angle gain required to offset synergistic deviations. By performing compensation feasibility envelope verification, the predicted increments are safely filtered and logically smoothed, ultimately outputting a corrected rolling process parameter set that accurately covers the effects of material heterogeneity.

[0102] In this solution, key factors affecting the steel plate rolling process are accurately identified through multi-dimensional data fusion and sensitivity analysis. Process parameters are optimized through a process compensation proxy model to improve forming accuracy and quality.

[0103] Based on the above technical solution, optionally, the roll-out adjustment module 106 is specifically used for: Based on the process control target data and the corrected rolling process parameter set, parameter redundancy elimination and global constraint boundary identification are performed in a control target-oriented manner to obtain the feasible domain solution space. Based on the feasible domain solution space, Pareto front optimization with objective conflict trade-off is performed to obtain a process parameter compensation scheme. Based on the process parameter compensation scheme and the preset physical response characteristic data of the rolling machine, kinematic feasibility verification and dynamic safety assessment are performed to obtain a benchmark execution sequence composed of optimized parameter commands and control gains. Based on the baseline execution sequence and the preset execution timing logic, semantic mapping and control logic embedding are performed to obtain the target process control strategy.

[0104] In this scheme, the feasible domain solution space is the range of all acceptable process parameters determined by eliminating redundancies and identifying global constraint boundaries within the framework of process control objectives and constraints. This space includes all effective parameter combinations that meet technical requirements and physical limitations, providing a reasonable boundary for the selection of process parameters.

[0105] The process parameter compensation scheme is a scheme for adjusting process parameters to compensate for deviations or errors that occur during the actual process execution, through optimization algorithms and multi-objective optimization calculations.

[0106] Predefined physical response characteristic data are physical parameters that are predefined during the equipment design phase or the accumulation of historical data to describe the response characteristics of equipment or processes. This type of data includes equipment operating limits, vibration characteristics, temperature response, mechanical properties, etc., and can be used to simulate the operating behavior of equipment under different operating conditions.

[0107] Optimized parameter instructions are equipment adjustment instructions calculated based on process parameter compensation schemes, used to guide equipment operation according to optimized parameters. These instructions mainly include adjustment requirements for control variables such as equipment operating point, speed, and pressure.

[0108] Control gain is the proportional gain in a closed-loop control system, used by the controller to amplify or reduce the error feedback signal. It directly affects the control system's response speed to deviations and its operational stability.

[0109] The baseline execution sequence is the optimal sequence of process operations determined after kinematic feasibility verification and dynamic safety assessment. It serves as a set of instructions executed in a predetermined logical order during process execution.

[0110] Preset execution sequence logic is a pre-defined sequence of actions and time relationships within the equipment control system, based on process requirements and operational procedures. It clarifies the sequential execution order of different process steps and the time interval between each step.

[0111] Based on the process control target data and the corrected rolling process parameter set, a multi-dimensional sensitivity screening technique is used to eliminate redundancy in the parameter space, identifying and filtering out weakly correlated variables that contribute very little to the final forming accuracy. Subsequently, a hyperplane boundary search technique is used to perform spatial intersection calculations on the working stroke, load limit, and product quality tolerance requirements of the rolling machine to clarify the global constraint boundary, thereby locking in the set of all parameter combinations that meet the physical and technical constraints, and finally determining the feasible solution space.

[0112] Within the feasible domain solution space, a multi-objective evolutionary optimization algorithm is used to address the conflict between production efficiency, roundness accuracy, and surface quality. The deviation of different parameter combinations in each objective dimension is calculated, and the optimal balance curve is plotted within the solution space. A weighted vector distance evaluation method is employed to select the optimal adjustment vector from the Pareto front that best matches the current offset, which is used to compensate for real-time detected machining errors. Based on this, a targeted process parameter compensation scheme is formulated.

[0113] By combining the process parameter compensation scheme with the preset physical response characteristic data of the rolling machine, discrete-time dynamic simulation is used to verify the kinematic feasibility of the compensation commands, ensuring that the displacement and pressure switching of the rollers do not exceed the response speed of the mechanical structure. Simultaneously, state-space stability analysis technology is introduced to evaluate the vibration response and temperature rise characteristics of the equipment under complex dynamic loads. By calculating the stability margin of the feedback loop, the optimal response coefficient and control step size are determined, and finally, a baseline execution sequence consisting of optimized parameter commands and control gains is synthesized.

[0114] Finally, using protocol instruction semantic mapping technology, the physical quantities in the baseline execution sequence are transformed into low-level logic code recognizable by the PLC. Based on the explicit sequence and time intervals in the preset execution timing logic, condition-triggered logic embedding technology is employed to directly integrate safety interlocks, action jumps, and interrupt protection instructions into the control flow. In this way, abstract operation instructions are transformed into a closed-loop execution set with strict time correlation and logical constraints, ultimately generating a target process control strategy that can directly drive the rolling machine to achieve automated and high-precision operation.

[0115] In this solution, by precisely optimizing process parameters, compensating for errors, and controlling gains, the accuracy and stability of the rolling process can be significantly improved, deviations caused by equipment limitations or external factors can be reduced, and the final product quality can be ensured to meet strict geometric and mechanical requirements.

[0116] See appendix Figure 2 , Figure 2 This is a second main structural block diagram of a fully automated production system for rolling steel tubes for gravure printing plates, according to an embodiment of the present invention. Figure 2 As shown, The system further includes a continuous monitoring module 107, which is used for: Continuously acquire real-time morphological data of the target gravure printing plate steel sheet during the rolling process, identify the current process based on the real-time morphological data, and perform error analysis and deviation extraction on the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to obtain morphological deviation data and process execution error information of the target gravure printing plate steel sheet during the rolling process. When the shape deviation data exceeds the preset accuracy standard, the rolling process parameter set is readjusted based on the shape deviation data and process execution error information. Reacquire the force and displacement response data of the target gravure printing plate steel plate, invert the material state data of the target gravure printing plate steel plate based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set. Based on the process control target data and the revised rolling process parameter set, multi-objective optimization calculations are performed again to generate a target process control strategy. The rolling machine is then adjusted based on the target process control strategy until the target gravure printing plate steel is rolled.

[0117] In this embodiment, the morphological data of the target gravure printing plate steel sheet is first acquired in real time, and the current process is identified. Error analysis is performed between the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to extract morphological deviation and process execution error information. If the morphological deviation meets the preset accuracy standard corresponding to the current process, monitoring continues; if the deviation exceeds the standard, the rolling process parameter set is adjusted based on the deviation and error information. Next, force and displacement response data are acquired again, the steel sheet material state is inverted, and regression analysis is performed in combination with the adjusted process parameters to obtain the corrected process parameters. Finally, multi-objective optimization calculations are performed based on the corrected parameters and process control target data to generate a process control strategy, and the rolling machine is adjusted in real time according to the strategy to ensure that the accuracy standard of the steel pipe obtained after the final rolling is met.

[0118] In this embodiment, by continuously monitoring the morphological deviations during the steel plate rolling process and adjusting the process parameters in real time, the forming quality of the steel pipe can be effectively guaranteed, ensuring that the preset accuracy standard is met, optimizing production efficiency, reducing error accumulation, and improving product consistency and reliability.

[0119] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effects of the present invention, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of the present invention.

[0120] Furthermore, the present invention also provides a fully automated production method for rolling steel tubes for gravure printing plates.

[0121] See appendix Figure 3 , Figure 3 This is a schematic flowchart of the main steps of a fully automated production method for rolling steel tubes for gravure printing plates according to an embodiment of the present invention. Figure 1As shown, a fully automated production method for rolling steel tubes for gravure printing plates in an embodiment of the present invention mainly includes the following steps S301-S306.

[0122] S301, Obtain the process requirements information of the target gravure printing plate steel plate, perform multi-dimensional data analysis and parameter derivation on the process requirements information, and obtain the preliminary process control parameter set of the target gravure printing plate steel plate.

[0123] S302, acquire historical process data that matches the feature parameters of the target gravure printing plate steel plate, perform time-series feature extraction and multi-dimensional data fusion based on the historical process data, extract sensitive key control variables in the steel plate rolling process, and optimize the preliminary process control parameter set based on the sensitive key control variables to obtain the rolling process parameter set.

[0124] S303, based on the rolling process parameter set, the rolling machine is controlled to perform a rolling operation on the target gravure printing plate steel plate, real-time morphological data of the target gravure printing plate steel plate during the rolling process is obtained, the current process is identified based on the real-time morphological data, and error analysis and deviation extraction are performed between the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to obtain morphological deviation data and process execution error information of the target gravure printing plate steel plate during the rolling process.

[0125] S304, If the shape deviation data exceeds the preset accuracy standard corresponding to the current process, adjust the rolling process parameter set based on the shape deviation data and process execution error information.

[0126] S305, acquire the force and displacement response data of the target gravure printing plate steel plate, invert the material state data of the target gravure printing plate steel plate based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set.

[0127] S306: Obtain process control target data, perform multi-objective optimization calculations based on the process control target data and the corrected rolling process parameter set, generate target process control strategy, and adjust the rolling machine based on the target process control strategy.

[0128] Those skilled in the art will understand that all or part of the processes in the method of the above embodiment of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or system capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable storage medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0129] Furthermore, the present invention also provides an electronic device 400, including a processor 401, a memory 402, and a program or instructions stored in the memory 402 and executable on the processor 401. When the program or instructions are executed by the processor 401, they implement the various processes of the above-described fully automated production method embodiment for rolling steel pipes for gravure printing plates, and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0130] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0131] Furthermore, the present invention also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to the present invention, the computer-readable storage medium can be configured to store a program for performing a fully automated production method for rolling steel tubes for gravure printing plates according to the above-described method embodiments. This program can be loaded and run by a processor to implement the above-described fully automated production method for rolling steel tubes for gravure printing plates. For ease of explanation, only the parts related to the embodiments of the present invention are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of the present invention. The computer-readable storage medium can be a storage system device comprising various electronic devices. Optionally, in the embodiments of the present invention, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0132] Furthermore, it should be understood that since the various modules are only provided to illustrate the functional units of the system of the present invention, the physical devices corresponding to these modules may be the processor itself, or a part of the processor's software, hardware, or a combination of software and hardware. Therefore, the number of modules shown in the figures is merely illustrative.

[0133] Those skilled in the art will understand that the various modules in the system can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principles of the present invention; therefore, the technical solutions after splitting or merging will fall within the protection scope of the present invention.

[0134] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A fully automated production system for rolling steel tubes for gravure printing plates, characterized in that... The system includes: The process requirements analysis module is used to obtain the process requirements information of the target gravure printing plate steel, perform multi-dimensional data analysis and parameter derivation on the process requirements information, and obtain the preliminary process control parameter set of the target gravure printing plate steel. The process parameter determination module is used to acquire historical process data that matches the feature parameters of the target gravure printing plate steel sheet, perform time-series feature extraction and multi-dimensional data fusion based on the historical process data, extract sensitive key control variables in the steel sheet rolling process, and optimize the preliminary process control parameter set based on the sensitive key control variables to obtain the rolling process parameter set; The deviation determination module is used to control the rolling machine to perform a rolling operation on the target gravure printing plate steel sheet based on the rolling process parameter set, acquire real-time morphological data of the target gravure printing plate steel sheet during the rolling process, identify the current process based on the real-time morphological data, and perform error analysis and deviation extraction on the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to obtain the morphological deviation data and process execution error information of the target gravure printing plate steel sheet during the rolling process; The adjustment module is used to adjust the rolling process parameter set based on the shape deviation data and process execution error information if the shape deviation data exceeds the preset accuracy standard corresponding to the current process. The process parameter correction module is used to acquire the force and displacement response data of the target gravure printing plate steel sheet, invert the material state data of the target gravure printing plate steel sheet based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set; The rolling adjustment module is used to acquire process control target data, perform multi-objective optimization calculations based on the process control target data and the corrected rolling process parameter set, generate target process control strategies, and adjust the rolling machine based on the target process control strategies.

2. The fully automated production system for rolling steel tubes for gravure printing plates according to claim 1, characterized in that... The process requirements analysis module is specifically used for: Multidimensional data parsing and parameter initialization are performed on the process requirements information to obtain a preliminary process requirements dataset; the preliminary process requirements dataset includes steel plate material property data and steel plate specification data; Based on the material property data of the steel plate, mechanical property analysis was performed to derive the plastic deformation resistance data and springback characteristic parameters of the target gravure printing plate steel plate; Based on the steel plate specification data, the shape and dimensional accuracy during the rolling process are preset to obtain a dataset of accuracy requirements; Based on preset equipment constraint data, accuracy requirement dataset, plastic deformation resistance data, and springback characteristic parameters, multi-pass deformation amount allocation and parameter integration are performed, and springback compensation correction is performed based on preset springback prediction model to obtain the preliminary process control parameter set of the target gravure printing plate steel plate.

3. The fully automated production system for rolling steel tubes for gravure printing plates according to claim 1, characterized in that... The process parameter determination module is specifically used for: Extract dynamic evolution features from historical process data to form a historical time-series feature dataset of the steel plate rolling process; Feature engineering and information fusion were performed on historical time-series feature datasets to obtain a comprehensive historical dataset of the steel plate rolling process. High-dimensional feature importance assessment was performed based on historical comprehensive datasets to extract sensitive key control variables in the steel plate rolling process; Based on sensitive key control variables, a multivariate coupled correlation analysis was conducted to obtain a set of factors affecting roll quality; Based on the set of factors affecting the quality of rolling, the preliminary set of process control parameters is optimized and adjusted in multiple objectives to obtain the rolling process parameter set.

4. The fully automated production system for rolling steel tubes for gravure printing plates according to claim 1, characterized in that... The deviation determination module is specifically used for: Based on the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process, spatial registration and difference calculation are performed to obtain the morphological deviation data of the target gravure printing plate steel plate during the rolling process; Error feature extraction and deviation analysis are performed based on morphological deviation data to obtain the spatial distribution pattern of morphological errors; Based on the spatial distribution pattern and the preset error propagation mechanism, source tracing analysis and evolutionary deduction are performed to obtain the process execution error information during the rolling process of the target gravure printing plate steel plate.

5. A fully automated production system for rolling steel tubes for gravure printing plates according to claim 1, characterized in that... The process parameter correction module is specifically used for: Based on the material state data, rheological features reflecting the material's rheological behavior and elastoplastic features reflecting the material's elastoplastic properties are extracted respectively. The rheological features, elastoplastic features, and the adjusted rolling process parameter set are subjected to cross-domain data alignment and modal fusion to construct a standardized dataset with enhanced features. A global sensitivity analysis was performed based on the standardized dataset to identify the contribution factors of rheological characteristics to the rolling execution deviation; Based on the aforementioned contribution factors and the adjusted set of rolling process parameters, a nonlinear coupling mechanism analysis was performed to obtain data on the synergistic effect and interactive influence of the material's elastic-plastic state and the rolling process parameters. Based on the synergistic effect data and interactive influence data, a process compensation proxy model is constructed. Based on the process compensation proxy model, incremental prediction and compensation analysis are performed on the adjusted rolling process parameter set to obtain the corrected rolling process parameter set.

6. The fully automated production system for rolling steel tubes for gravure printing plates according to claim 1, characterized in that... The roll-out adjustment module is specifically used for: Based on the process control target data and the corrected rolling process parameter set, parameter redundancy elimination and global constraint boundary identification are performed in a control target-oriented manner to obtain the feasible domain solution space; Based on the feasible domain solution space, Pareto front optimization with objective conflict trade-offs is performed to obtain a process parameter compensation scheme; Based on the aforementioned process parameter compensation scheme and the preset physical response characteristic data of the rolling machine, kinematic feasibility verification and dynamic safety assessment are performed to obtain a baseline execution sequence composed of optimized parameter commands and control gains; Based on the baseline execution sequence and the preset execution timing logic, semantic mapping and control logic embedding are performed to obtain the target process control strategy.

7. The fully automated production system for rolling steel tubes for gravure printing plates according to claim 1, characterized in that... The system further includes a continuous monitoring module, which is used for: The system continuously acquires real-time morphological data of the target gravure printing plate steel sheet during the rolling process. Based on the real-time morphological data, the current process is identified, and error analysis and deviation extraction are performed between the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to obtain morphological deviation data and process execution error information of the target gravure printing plate steel sheet during the rolling process. When the shape deviation data exceeds the preset accuracy standard corresponding to the current process, the rolling process parameter set is readjusted based on the shape deviation data and process execution error information. Reacquire the force and displacement response data of the target gravure printing plate steel sheet, invert the material state data of the target gravure printing plate steel sheet based on the force and displacement response data, and perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set; Multi-objective optimization calculations are performed again based on the process control target data and the revised rolling process parameter set to generate a target process control strategy. The rolling machine is then adjusted based on the target process control strategy until the target gravure printing plate steel is rolled into a circle.

8. A fully automated production method for rolling steel tubes for gravure printing plates, characterized in that... The method includes: Obtain the process requirements information of the target gravure printing plate steel, perform multi-dimensional data analysis and parameter derivation on the process requirements information, and obtain the preliminary process control parameter set of the target gravure printing plate steel; Historical process data matching the feature parameters of the target gravure printing plate steel is obtained. Based on the historical process data, time-series feature extraction and multi-dimensional data fusion are performed to extract sensitive key control variables in the steel plate rolling process. Based on the sensitive key control variables, the preliminary process control parameter set is optimized to obtain the rolling process parameter set. The rolling machine is controlled based on the rolling process parameter set to perform rolling operation on the target gravure printing plate steel sheet. Real-time morphological data of the target gravure printing plate steel sheet is obtained during the rolling process. The current process is identified based on the real-time morphological data, and error analysis and deviation extraction are performed between the real-time morphological data and the theoretical forming trajectory reference data corresponding to the current process to obtain the morphological deviation data and process execution error information of the target gravure printing plate steel sheet during the rolling process. If the shape deviation data exceeds the preset accuracy standard corresponding to the current process, adjust the rolling process parameter set based on the shape deviation data and process execution error information. Obtain the force and displacement response data of the target gravure printing plate steel sheet; invert the material state data of the target gravure printing plate steel sheet based on the force and displacement response data; perform correlation modeling and regression analysis based on the material state data and the adjusted rolling process parameter set to obtain the corrected rolling process parameter set. The process control target data is obtained, and multi-objective optimization calculations are performed based on the process control target data and the corrected rolling process parameter set to generate the target process control strategy. The rolling machine is then adjusted based on the target process control strategy.

9. An electronic device, comprising a processor, a memory, and a program or instructions stored in the memory and executable on the processor, characterized in that... The program or instructions are adapted to be loaded and run by the processor to perform a fully automated production method for rolling steel tubes for gravure printing plates as described in claim 8.

10. A computer-readable storage medium storing a plurality of program codes, characterized in that... The program code is adapted to be loaded and run by a processor to perform a fully automated production method for rolling steel tubes for gravure printing plates as described in claim 8.

Images