Hot continuous rolling laminar cooling control method and system
By using a prediction model that integrates multi-source data and performs dynamic correction, the problems of insufficient sensing and weak self-correction capability in the laminar cooling control of hot continuous rolling have been solved, achieving high-precision coiling temperature control and improved yield.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-11
- Publication Date
- 2026-03-31
AI Technical Summary
Existing hot continuous rolling laminar flow cooling control technology suffers from limited sensing dimensions and insufficient self-correction capabilities when facing complex and ever-changing industrial scenarios, resulting in uneven coiling temperature and reduced yield.
A prediction model based on multi-source data fusion, combined with gradient boosting decision trees and attention mechanisms, is adopted. By collecting multi-source process parameters in real time, a multi-source data input vector is constructed, and a feedforward model is used for dynamic correction to achieve precise control of the number of valves opened.
It improves the adaptability and accuracy of laminar flow cooling control, enhances the uniformity of winding temperature and yield, and strengthens the stability and reliability of the system.
Smart Images
Figure CN121755554A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of hot rolling process control technology, and in particular to a hot continuous rolling laminar flow cooling control method and system. Background Technology
[0002] The laminar flow cooling process in hot strip rolling is a crucial step in controlling the final mechanical properties and microstructure of the strip. Its core objective is to rapidly cool the strip from the high temperature at the finishing mill exit to the target coiling temperature within tens of seconds through precise control of the cooling water jet. Current developments in laminar flow cooling control technology mainly focus on the following directions:
[0003] Dynamic process control: Existing technologies have addressed the impact of dynamic factors such as rolling speed on cooling efficiency. For example, some literature proposes dynamically dividing cooling control sections based on strip speed to address variations in strip residence time in the cooling zone caused by speed changes.
[0004] Model and parameter optimization: To improve the model's adaptability, researchers focused on optimizing the model itself. This included introducing fuzzy identification methods to model key parameters such as the water-cooling heat transfer coefficient.
[0005] Equipment and Standards Innovation: At the hardware level, innovative equipment adjustments and standardized operations ensure cooling uniformity. For example, the "pentagonal positioning method" is used to precisely control the side spray parameters, effectively solving the problem of uneven cooling caused by "water accumulation" on the strip surface.
[0006] However, existing control models still have the following limitations when dealing with complex and ever-changing real-world industrial scenarios: Limited Perception Dimensions: Most control models still heavily rely on static process parameters such as steel grade and target thickness, lacking in-depth integration and utilization of multi-source and heterogeneous data from the production site. For example, existing models often fail to adequately consider equipment status parameters that directly affect heat exchange efficiency, such as cooling water temperature, water pressure fluctuations, and the availability of key valve positions, as well as the production context information of the preceding strip steel, resulting in insufficient perception of the overall production conditions.
[0007] Lagging correction mechanisms: Many models lack effective real-time self-correction capabilities. When changes in rolling speed, fluctuations in final rolling temperature, or changes in equipment status cause the cooling effect to deviate from expectations, the system struggles to make timely and accurate adjustments. This easily leads to a decrease in coiling temperature hit rate, especially with frequent problems such as excessive temperature deviations at the head and tail of the strip, affecting the uniformity of product performance and yield.
[0008] Limited adaptability: Production conditions are characterized by long-term, slow changes (such as equipment aging and water quality changes). Existing technologies are generally weak in their ability to continuously adapt to such long-term changes, and their performance gradually declines over time once the model is fixed. Summary of the Invention
[0009] This invention provides a hot strip mill laminar flow cooling control method and system to solve the problem of decreased control performance of existing hot strip mill laminar flow cooling temperature prediction models when key process parameters such as rolling speed and final rolling temperature fluctuate.
[0010] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, the present invention provides a method for controlling laminar flow cooling in hot continuous rolling, comprising: Collect multi-source process parameters in the laminar flow cooling process of hot continuous rolling and construct a multi-source data input vector; The multi-source data input vector is input into a preset prediction model to obtain the predicted value of the number of valves opened; wherein, the prediction model takes the multi-source data input vector as input and the number of valves opened for laminar cooling as output; Based on the process parameters of the preceding strip steel, the valve opening quantity correction amount is calculated in real time through a preset feedforward model; The predicted number of valves to be opened is superimposed with the corrected number of valves to be opened to obtain the final number of valves to be opened, and the final number of valves to be opened is sent to the control actuator of the hot continuous rolling laminar flow cooling process.
[0011] Furthermore, the acquisition of multi-source process parameters in the hot continuous rolling laminar flow cooling process, and the construction of a multi-source data input vector, includes: Collect multi-source process parameters in the laminar flow cooling process of hot continuous rolling; wherein, the multi-source process parameters include static process parameters, real-time process parameters, equipment status parameters, and historical production sequence parameters; The collected multi-source process parameters are preprocessed, and feature engineering is performed on the preprocessed multi-source process parameters to construct a multi-source data input vector that integrates static process parameters, real-time process parameters, equipment status parameters, and historical production sequence parameters.
[0012] Furthermore, the static process parameters include the steel grade of the current steel coil, the target thickness, and the target coiling temperature; The real-time process parameters include the current rolling speed and final rolling temperature of the steel coil; The equipment status parameters include cooling water pressure, cooling water temperature, and valve health status. The historical production sequence parameters include the steel type, thickness, and coiling temperature control results of the preceding N strips; where N is a preset integer value greater than 0.
[0013] Furthermore, the preprocessing of the collected multi-source process parameters includes: Data alignment and cleaning are performed on the collected multi-source process parameters.
[0014] Furthermore, the prediction model is a gradient boosting decision tree model with an added attention mechanism.
[0015] Furthermore, the feedforward model takes the process parameters of the preceding strip as input and the valve opening quantity correction of the current strip as output; The process parameters of the preceding strip include: the coiling temperature deviation of the previous strip, the process similarity between the previous strip and the current strip, and the predicted number of valves to be opened for the previous strip.
[0016] Furthermore, the feedforward model employs an online sequence learning algorithm, the mathematical expression of which is: ΔU(k) = K(k)·e(k-1) + λ·ΔU(k-1) + β·S(k-1)·U_pred(k-1) Where ΔU(k) is the correction amount at the current time; e(k-1) is the coiling temperature deviation of the previous piece of steel; and K(k) is the time-varying correction gain. ΔU(k-1) is the forgetting factor for historical corrections; ΔU(k-1) is the correction for the number of valves opened in the previous strip; S(k-1) is the process similarity between the previous strip and the current strip; U_pred(k-1) is the predicted value for the number of valves opened in the previous strip; β is the preset prediction value weighting coefficient.
[0017] Furthermore, The value is updated in real time using the recursive least squares method.
[0018] Furthermore, the hot continuous rolling laminar flow cooling control method also includes: periodically using multi-source process parameters generated during historical production processes to perform incremental learning and version iteration on the prediction model and the feedforward model.
[0019] On the other hand, the present invention also provides a hot continuous rolling laminar flow cooling control system, comprising: The multi-source data integration module is used to collect multi-source process parameters in the laminar cooling process of hot continuous rolling and construct a multi-source data input vector. The intelligent decision engine module is used for: The multi-source data input vector is input into a preset prediction model to obtain the predicted value of the number of valves opened; wherein, the prediction model takes the multi-source data input vector as input and the number of valves opened for laminar cooling as output; Based on the process parameters of the preceding strip steel, the valve opening quantity correction amount is calculated in real time through a preset feedforward model; The control command distribution module is used to superimpose the predicted number of valves to be opened with the corrected number of valves to be opened to obtain the final number of valves to be opened, and then send the final number of valves to the control execution mechanism of the hot continuous rolling laminar flow cooling process.
[0020] In another aspect, the present invention also provides an electronic device comprising a processor and a memory; wherein the memory stores at least one instruction, which is loaded and executed by the processor to implement the above-described method.
[0021] In another aspect, the present invention also provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the above method.
[0022] The beneficial effects of the technical solution provided by this invention include at least the following: 1. Multiphysics field coupling modeling theory: This invention is the first to unify multiple physical processes such as material phase transformation dynamics, unsteady heat transfer, and equipment performance degradation under a data-driven framework, and establishes a new prediction model that deeply integrates mechanism and data.
[0023] 2. Adaptive System Identification Algorithm: This invention proposes a dynamic correction theory based on incremental learning, which tracks the changes in the dynamic characteristics of the system in real time through online parameter estimation, thus solving the problem that traditional fixed parameter controllers are difficult to adapt to operating condition drift.
[0024] 3. Robust Optimized Control Architecture: This invention constructs a predictive control framework that considers multiple constraints, which significantly improves the stability and reliability of the system while ensuring control accuracy.
[0025] 4. Industrial Artificial Intelligence Paradigm: This invention establishes a new paradigm of industrial intelligent control, namely "physically guided machine learning + online adaptive correction", providing a replicable and scalable technical route for the control of complex industrial processes. Attached Figure Description
[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0027] Figure 1 This is a schematic flowchart of the hot continuous rolling laminar flow cooling control method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the multi-source input data structure provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the principle of the dynamic correction steps provided in this embodiment of the invention; Figure 4 This is an architecture diagram of the hot continuous rolling laminar flow cooling control system provided in an embodiment of the present invention; Figure 5This is a system block diagram of the electronic device provided in the embodiments of the present invention. Detailed Implementation
[0028] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
[0029] First, it should be noted that in the embodiments of the present invention, the words "exemplarily," "for example," etc., are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the term "exemplarily" is intended to present the concept in a specific manner. Furthermore, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either one or the other.
[0030] First Embodiment To address the problem of insufficient laminar flow cooling control accuracy in existing technologies due to fluctuations in operating conditions, changes in equipment efficiency, and lack of dynamic modeling, this embodiment provides a hot continuous rolling laminar flow cooling control method based on multi-source data fusion, featuring high-dimensional perception and dynamic feedforward self-correction capabilities based on historical experience, thereby improving the control robustness of the control system under complex operating conditions. This method can be implemented by electronic devices, such as terminals or servers. The execution flow of this method includes the following steps:
[0031] S1, collect multi-source process parameters in the laminar flow cooling process of hot continuous rolling and construct a multi-source data input vector; Specifically, in this embodiment, the implementation process of S1 is as follows: S11, Collect multi-source process parameters in the laminar flow cooling process of hot continuous rolling; wherein, the multi-source process parameters include static process parameters, real-time process parameters, equipment status parameters, and historical production sequence parameters; Furthermore, static process parameters include the steel grade, target thickness, and target coiling temperature of the current steel coil; real-time process parameters include the rolling speed and final rolling temperature of the current steel coil; equipment status parameters include cooling water pressure, cooling water temperature, and valve health status; historical production sequence parameters are used to characterize the dynamic operating conditions of the production line, including the steel grade, thickness, and coiling temperature control results of the previous N strips; where N is an integer greater than 0.
[0032] S12, preprocessing the collected multi-source process parameters, including: unifying data with different sampling frequencies and transmission delays to the same time base through a time alignment algorithm, and performing data cleaning.
[0033] S13. Perform feature engineering on the preprocessed multi-source process parameters to construct a multi-source data input vector that integrates static process parameters, real-time process parameters, equipment status parameters, and historical production sequence parameters.
[0034] Specifically, this embodiment constructs a feature engineering system based on a combination of physical mechanisms and data-driven approaches, including: 1) Construct equivalent cooling intensity characteristics based on the principles of heat transfer, taking into account the effects of water pressure, water temperature, and valve health status on the heat transfer coefficient; 2) Construct a dynamic heat accumulation factor to quantify the thermal inertia effect of roller conveyors and water tanks based on historical production sequences and equipment operating status; 3) Introduce a process stability index to reflect the dynamic characteristics of the cooling process through the fluctuation characteristics of the rolling speed.
[0035] S2, input the multi-source data input vector into the preset prediction model to obtain the predicted value of the number of valves opened; wherein, the prediction model takes the multi-source data input vector as input and the number of valves opened for laminar cooling as output; In this embodiment, the prediction model adopts a hybrid architecture combining gradient boosting decision trees and attention mechanisms, wherein: 1) The gradient boosting decision tree module is responsible for learning the deterministic mapping relationship between static process parameters and equipment state parameters; 2) The attention mechanism module is responsible for dynamically weighting the influence of real-time process parameters, paying particular attention to the temporal variation patterns of rolling speed and final rolling temperature; 3) The model enhances its robustness to abnormal operating conditions through adversarial training, ensuring the stability of predictions under conditions such as sudden speed changes and abnormal temperatures.
[0036] S3, based on the process parameters of the preceding strip steel, calculates the valve opening quantity correction amount in real time through a preset feedforward model; In this embodiment, the process parameters of the preceding strip include: the coiling temperature deviation of the previous strip, the process similarity between the previous strip and the current strip, and the predicted number of valves to be opened for the previous strip. Based on this, this embodiment constructs a feedback correction mechanism based on the initial prediction of the number of valves to be opened. Based on the control deviation and process similarity of the preceding strip, a lightweight feedforward model calculates the dynamic valve number correction amount in real time; thereby generating the final valve number command and achieving dynamic adaptive adjustment of the cooling process. This model takes the coiling temperature deviation, process similarity, and initial prediction values of the previous strip as input to perform rapid reasoning adapted to the production rhythm.
[0037] Specifically, in this embodiment, the correction model employs an online sequence learning algorithm, and its mathematical model is as follows: ΔU(k) = K(k)·e(k-1) + λ·ΔU(k-1) + β·S(k-1)·U_pred(k-1) in, Here, e(k-1) represents the current correction value, e(k-1) represents the coiling temperature deviation of the previous piece of steel, and K(k) represents the time-varying correction gain. ΔU(k-1) is the forgetting factor for historical corrections; ΔU(k-1) is the correction for the number of valves opened in the previous strip; S(k-1) is the process similarity between the previous strip and the current strip (normalized value between 0 and 1); U_pred(k-1) is the predicted value for the number of valves opened in the previous strip; β is the weighting coefficient for the predicted value.
[0038] Process similarity can be calculated as follows: S(k-1) = 1 - \frac{|T_{entry}(k) - T_{entry}(k-1)| + |Thickness(k) - Thickness(k-1)|}{Normalization Factor}.
[0039] λ = λ_0 · exp(-α·|e(k-1)|) + λ_min; where λ_0 is the basic forgetting factor (recommended 0.6-0.8); λ_min is the minimum forgetting factor (recommended 0.2-0.3); α is the preset attenuation coefficient, determined according to the historical error distribution; when the error is large, λ decreases to reduce the influence of historical corrections; when the error is small, λ increases to maintain the continuity of correction.
[0040] The correction gain K(k) is updated in real time using the recursive least squares method, enabling the system to possess progressive self-learning capabilities.
[0041]
[0042] in, For learning rate, It is the covariance matrix; This is the correction gain from the previous time step; Let be the covariance matrix of the previous time step.
[0043] S4, the predicted number of valves to be opened is superimposed with the corrected number of valves to be opened to obtain the final number of valves to be opened, and the final number of valves to be opened is sent to the control actuator of the hot continuous rolling laminar flow cooling process.
[0044] Furthermore, this embodiment comprehensively considers control accuracy, energy consumption, and stability requirements, and constructs a constrained optimization problem:
[0045]
[0046] in, To predict the winding temperature, For the final number of valves opened, , These are the weighting coefficients. The target winding temperature; The nominal number of valves opened; Minimum number of valves to open; This represents the maximum number of valves that can be opened. This is the amount of correction for a single operation; This is the maximum permissible correction amount.
[0047] By solving this optimization problem, control commands that balance accuracy and stability are generated.
[0048] Furthermore, the hot continuous rolling laminar flow cooling control method also includes: periodically using multi-source process parameters generated during historical production processes to perform incremental learning and version iteration on the prediction model and feedforward model.
[0049] Through the above methods, this embodiment constructs an intelligent hot continuous rolling laminar flow cooling control method that combines feedforward predictive and feedback error correction capabilities, effectively solving the dynamic fluctuation problem in the laminar flow cooling process.
[0050] Second Embodiment This embodiment, in conjunction with the accompanying drawings, provides a more detailed description of the implementation process of the method of the present invention. Specifically, the hot continuous rolling laminar flow cooling control method mainly includes the following steps: S1: Multi-source heterogeneous data acquisition and collaborative characterization steps; It should be noted that, for example Figure 1 As shown, the core process of this invention begins with the S1 data acquisition step. For example... Figure 2 As shown, in this embodiment, the multi-source data specifically includes:
[0051] 1) Static process parameters: In addition to conventional parameters, material phase transformation kinetics characteristics are introduced, and the critical transformation point temperature of the CCT curve is calculated based on the chemical composition of the steel, providing a theoretical basis for cooling strategies in different phase transformation zones.
[0052] 2) Real-time process parameters: Based on unsteady-state heat transfer theory, a cooling time correction model considering rapid velocity changes is established:
[0053] in, For effective cooling time, The length of the cooling zone. Instantaneous rolling speed, This is the sensitivity coefficient for speed changes. This model significantly improves control accuracy under speed fluctuation conditions.
[0054] 3) Equipment status parameters: obtained from the cooling system PLC, including main pipe water pressure, water tank level / temperature, and valve health status indicators provided by the equipment inspection system. A valve performance degradation model is constructed to predict the degree of degradation in actual heat exchange capacity based on parameters such as valve actuation frequency and operating water temperature.
[0055]
[0056] in, For the current valve efficiency, For initial performance, For the number of actions, The attenuation coefficient is... This is the function that affects water temperature.
[0057] 4) Historical context information: Obtain the steel grade and production sequence of the previous piece of steel from the Manufacturing Execution System (MES), establish a thermal state transfer model of the equipment, and obtain the temperature field evolution of equipment such as roller conveyors and guides by discretely solving the heat conduction equation, so as to accurately quantify the heat accumulation effect.
[0058] S2: Initial prediction step based on multimodal fusion; In the initial prediction step, the multi-source data, after preprocessing (such as normalization and outlier filtering), forms a unified feature vector. This vector is then input into an offline-trained gradient boosting decision tree model. This invention proposes a novel modeling method combining multi-scale feature extraction with physically guided machine learning:
[0059] 1) The model adopts a hierarchical attention mechanism, which automatically learns the importance weights of different process parameters at the feature level and captures the evolution of dynamic parameters such as speed and temperature at the time level.
[0060] 2) Introduce physical consistency constraints to ensure that the model prediction results conform to the basic laws of thermodynamics, such as energy conservation and monotonically decreasing temperature.
[0061] 3) Improve the model's generalization ability on out-of-distribution samples by using adversarial regularization techniques, specifically by adding the following to the loss function:
[0062] in, Mean square error, To counteract the prediction error of adversarial samples; To weigh the parameters.
[0063] The learning objective of this model is to find a complex mapping relationship from multiple input sources to the optimal number of valves to open, based on massive historical data. In particular, in this embodiment, since rolling speed and final rolling temperature are strong features involved in the learning process, the model can automatically adjust its predicted output when these variables fluctuate, which is fundamentally different from traditional models that rely on fixed parameters.
[0064] S3: Dynamic correction steps based on incremental learning; In the dynamic correction step, such as Figure 3 As shown, this embodiment establishes an intelligent feedback mechanism based on incremental system identification: 1) The system records three key data points for the previous piece of steel (N-1): its initial predicted valve opening amount, the actual valve opening amount, and the final coiling temperature deviation ΔT.
[0065] 2) Input these data into a simple linear correction model: correction amount The coefficient K can be updated online using recursive least squares. This model learns the relationship between temperature deviation and the required valve adjustment.
[0066] 3) The correction value output by the model will be applied to the initial predicted value of the current block steel (N).
[0067] The calibration model is essentially an ARX model with time-varying parameters:
[0068] Parameter vector Online updates via recursive least squares with a forgetting factor:
[0069] in, Forgetting factor (0 < ≤1), As a regression vector, this algorithm enables the system to track slow changes in the dynamic characteristics of the process.
[0070] S4: Control output steps for multi-objective optimization; In the control output step, this embodiment employs a robust model predictive control framework to handle uncertainties:
[0071]
[0072]
[0073]
[0074] Where H represents the prediction time domain, and Q and R are weight matrices. For a bounded disturbance, It is a composite system model based on an initial prediction model and a correction model.
[0075] Finally, the corrected final valve opening quantity command is sent to the PLC in the laminar flow cooling zone through the control network, and the PLC drives the corresponding valve action.
[0076] Third Embodiment This embodiment provides a hot continuous rolling laminar flow cooling control system, such as Figure 4 As shown, the system adopts an architecture of data source layer, middleware layer, and control layer, and includes the following modules: The multi-source data integration module is deployed at the data source layer and is used to collect multi-source process parameters from the L1 and L2 systems and sensor networks in real time during the laminar flow cooling process of hot continuous rolling and to construct a multi-source data input vector. The intelligent decision engine module, deployed in the middle layer, is equipped with a prediction model and a correction model, and is used for: The multi-source data input vector is input into the preset prediction model to obtain the predicted value of the number of valves to be opened; at the same time, based on the process parameters of the preceding strip steel, the correction amount of the number of valves to be opened is calculated in real time through the preset feedforward model. The control command distribution module, deployed in the control layer, is used to superimpose the predicted valve opening number with the corrected valve opening number to obtain the final valve opening number, and then send it to the control execution mechanism of the hot continuous rolling laminar flow cooling process.
[0077] Furthermore, the system also includes a model self-updating module, which is used to periodically perform incremental learning and version iteration on the machine learning model and the lightweight calibration model using historical production data.
[0078] It should be noted that the hot rolling laminar flow cooling control system of this embodiment corresponds to the hot rolling laminar flow cooling control method of the first embodiment described above; the functions implemented by each functional module in the hot rolling laminar flow cooling control system of this embodiment correspond one-to-one with the process steps in the hot rolling laminar flow cooling control method of the first embodiment described above; therefore, they will not be described again here.
[0079] Third Embodiment This embodiment provides an electronic device, such as... Figure 5 As shown, the electronic device includes a processor and a memory; wherein the processor and the memory can be connected via a communication bus; the memory stores at least one instruction, which is loaded and executed by the processor to implement the method of the first embodiment described above. Furthermore, the electronic device may also include a transceiver, the processor and the transceiver can be connected via a communication bus, and the transceiver is used to communicate with other devices.
[0080] Below, in conjunction with Figure 5 A detailed introduction to each component of this electronic device is provided below: The processor is the control center of the electronic device. The electronic device may include multiple processors, each of which can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The term "processor" can refer to a single processor or a collective term for multiple processing elements. For example, a processor can be one or more central processing units (CPUs), other general-purpose processors, application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), one or more field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor can perform various functions of the electronic device by running or executing software programs stored in memory and by calling data stored in memory.
[0081] In a specific implementation, as one example, the processor may include one or more CPUs, for example... Figure 5 CPU0 and CPU1 shown are, of course, merely illustrative examples.
[0082] The memory is used to store the software program that executes the solution of the present invention, and the processor controls its execution. For specific implementation methods, please refer to the above method embodiments, which will not be repeated here.
[0083] Optionally, the memory may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory may be integrated with the processor or exist independently, and may be accessed through the interface circuit of the electronic device ( Figure 5 (Not shown in the image) is coupled to the processor; however, this embodiment of the invention does not impose specific limitations on this.
[0084] The transceiver may include a receiver and a transmitter. Figure 5 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function. The transceiver can be integrated with the processor or exist independently, and can be connected through the interface circuit of the electronic device (…). Figure 5 (Not shown in the image) is coupled to the processor, and this embodiment of the invention does not specifically limit this.
[0085] In addition, it should be noted that, Figure 5 The structure of the electronic device shown is not intended to limit the device. Actual devices may include more or fewer components than shown, or combine certain components, or have different component arrangements. Furthermore, the technical effects achieved by this electronic device when performing the method of the first embodiment described above can be referenced to the technical effects described in the first embodiment; therefore, they will not be repeated here.
[0086] Fourth embodiment This embodiment provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the method of the first embodiment described above. The computer-readable storage medium may be a ROM, random access memory, CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc. The instruction stored therein can be loaded and executed by a processor in a terminal.
[0087] Furthermore, it should be noted that the present invention can be provided as a method, apparatus, or computer program product. Therefore, embodiments of the present invention can take the form of a completely or partially hardware embodiment, a completely or partially software embodiment, or an embodiment combining software and hardware aspects. Moreover, when implemented in software, embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any usable medium accessible to a computer or a data storage device such as a server or data center containing one or more sets of usable media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive (SSD).
[0088] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal equipment to cause a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0090] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element. Furthermore, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Additionally, the character " / " in this text generally indicates an "or" relationship between the preceding and following objects, but it can also indicate an "AND / OR" relationship. Please refer to the context for specific interpretations. "At least one" refers to one or more items, while "more than" refers to two or more items. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can be represented as: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.
[0091] Furthermore, it is understood that in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0092] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.
[0093] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of functional modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms. Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs. Additionally, the functional units in the various embodiments of this invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0094] If the method is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0095] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention. It should be pointed out that although preferred embodiments of the present invention have been described, those skilled in the art, once they understand the basic inventive concept of the present invention, can make several improvements and modifications without departing from the principles described herein. These improvements and modifications should also be considered within the scope of protection of the present invention. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.
Claims
1. A hot strip continuous casting and rolling laminar cooling control method, characterized by, The application relates to a hot continuous rolling laminar cooling control method. The application comprises the following steps: Collecting multi-source process parameters in a hot continuous rolling laminar cooling process to construct a multi-source data input vector; Inputting the multi-source data input vector into a preset prediction model to obtain a valve opening quantity prediction value; wherein the prediction model takes the multi-source data input vector as input and takes the laminar cooling valve opening quantity as output; Based on the process parameters of a previous strip steel, a valve opening quantity correction quantity is calculated in real time through a preset feedforward model; 2. The hot continuous rolling laminar cooling control method as claimed in claim 1, characterized by, The valve opening quantity prediction value and the valve opening quantity correction quantity are superimposed to obtain a final valve opening quantity, and the final valve opening quantity is transmitted to a control execution mechanism of the hot continuous rolling laminar cooling process. The application comprises the following steps: Collecting multi-source process parameters in a hot continuous rolling laminar cooling process to construct a multi-source data input vector; 3. The hot strip flow cooling control method as recited in claim 2, characterized by, Collecting multi-source process parameters in the hot continuous rolling laminar cooling process; wherein the multi-source process parameters comprise static process parameters, real-time process parameters, equipment state parameters and historical production sequence parameters; The collected multi-source process parameters are preprocessed, and feature engineering is performed on the preprocessed multi-source process parameters to construct a multi-source data input vector which integrates the static process parameters, the real-time process parameters, the equipment state parameters and the historical production sequence parameters. The static process parameters comprise the steel grade, the target thickness and the target coiling temperature of the current steel coil; The real-time process parameters comprise the rolling speed and the finish rolling temperature of the current steel coil; 4. The hot strip flow cooling control method as recited in claim 2, characterized by, The equipment state parameters comprise the cooling water pressure, the cooling water temperature and the valve health state; The historical production sequence parameters comprise the steel grade, the thickness and the coiling temperature control result of the previous N blocks of strip steels; wherein N is a preset integer greater than 0.
5. The hot strip flow cooling control method as recited in claim 1, wherein The preprocessing of the collected multi-source process parameters comprises the following steps:
6. The hot strip flow cooling control method as recited in claim 1, wherein, Data alignment and data cleaning are performed on the collected multi-source process parameters. The prediction model is a gradient boosting decision tree model with an attention mechanism.
7. The hot strip flow cooling control method as recited in claim 1, wherein The feedforward model takes the process parameters of the previous strip steel as input and takes the valve opening quantity correction quantity of the current strip steel as output; The process parameters of the previous strip steel comprise the coiling temperature deviation of the previous strip steel, the process similarity between the previous strip steel and the current strip steel, and the valve opening quantity prediction value of the previous strip steel. Wherein, ΔU(k) is the current time correction amount; e(k-1) is the coiling temperature deviation of the previous strip; K(k) is the time-varying correction gain; is a forgetting factor of the historical correction amount; ΔU(k-1) is the valve opening quantity correction amount of the previous strip; S(k-1) is the process similarity degree between the previous strip and the current strip; U_pred(k-1) is the valve opening quantity prediction value of the previous strip; and β is a preset prediction value weight coefficient.
8. The hot strip flow cooling control method as recited in claim 7, characterized by, The values of the parameters are updated in real time by recursive least squares.
9. The hot strip flow cooling control method as recited in claim 1, wherein, The feedforward model adopts an online sequence learning algorithm, and the mathematical expression is as follows:
10. A hot strip flow cooling control system characterized by, Delta U (k) = K (k) * e (k-1) + Lambda * Delta U (k-1) + Beta * S (k-1) * U_pred (k-1); The hot continuous rolling laminar cooling control method further comprises the following steps: periodically performing incremental learning and version iteration on the prediction model and the feedforward model by using the multi-source process parameters generated in the historical production process. The application comprises the following steps: A multi-source data integration module is used to collect multi-source process parameters in a hot continuous rolling laminar cooling process to construct a multi-source data input vector; An intelligent decision engine module is used to: Input the multi-source data input vector into a preset prediction model to obtain a valve opening quantity prediction value; wherein the prediction model takes the multi-source data input vector as input and takes the laminar cooling valve opening quantity as output; Based on the process parameters of a previous strip steel, a valve opening quantity correction quantity is calculated in real time through a preset feedforward model; The valve opening quantity prediction value and the valve opening quantity correction quantity are superimposed to obtain a final valve opening quantity, and the final valve opening quantity is transmitted to a control execution mechanism of the hot continuous rolling laminar cooling process. The control instruction distribution module is configured to superimpose the open valve quantity prediction value and the open valve quantity correction value to obtain a final open valve quantity, and deliver the final open valve quantity to a control execution mechanism of the hot continuous rolling laminar cooling process.