Finish rolling inlet temperature prediction control method and device, electronic equipment and storage medium

Abstract

The application provides a finish rolling inlet temperature prediction control method and device, electronic equipment and storage medium, relates to the rolling technology field, and the method first constructs a temperature prediction control model based on historical data in the hot rolling cooling process of different plates. Then, according to the temperature data collected at the real temperature time and the output result of the temperature prediction control model, the parameters of the temperature prediction control model are dynamically adjusted. Based on the deviation between the temperature distribution predicted by the temperature prediction control model and the target temperature distribution, the parameters of the cooling system are adjusted to control the cooling temperature at the finish rolling inlet. The above method solves the technical problem of low cooling temperature control precision in the hot rolling production process, and achieves the technical effects of improving the finish rolling inlet temperature stability, improving the cooling efficiency and product quality stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of steel rolling, in particular to a finishing rolling entry temperature prediction control method and device, an electronic device and a storage medium. BACKGROUND

[0002] Hot rolling production process is one of the important processing links in the steel industry, mainly used for rolling the heated billet into plate with certain shape and size. In the hot rolling process, the finishing rolling process has a decisive influence on the quality of the final product, and the finishing rolling entry temperature control is a key parameter to ensure product quality and production stability.

[0003] Modern hot rolling production line generally uses air-water atomizing cooling system to adjust the temperature of the steel strip at the finishing rolling entry to realize temperature control. The typical air-water atomizing cooling system is composed of a fan, an air duct, an array of nozzles, cooling water and a temperature sensor. The cooling air volume and speed are controlled by adjusting the fan speed and damper opening, and the nozzles spray the atomized medium formed by mixing cooling water and compressed air to the surface of the steel strip. The temperature sensor collects the surface temperature of the steel strip in real time and feeds back the data to the control system to realize closed-loop regulation.

[0004] However, in actual production process, due to the change of steel strip running speed, mechanical inertia of cooling system and environmental temperature fluctuation and other factors, the existing air-water atomizing cooling system often has the problem of insufficient temperature control precision, which directly affects the rolling force distribution and product size precision, and in severe cases, it will cause surface quality defects of the strip and abnormal load of the rolling mill. SUMMARY

[0005] The purpose of the embodiments of the present application is to provide a finishing rolling entry temperature prediction control method, device, electronic device and storage medium, to solve the technical problem of low cooling temperature control precision in the hot rolling production process in the prior art.

[0006] In a first aspect, the embodiments of the present application provide a finishing rolling entry temperature prediction control method, which comprises:

[0007] A temperature prediction control model is constructed based on historical data in the hot rolling cooling process of different plates, and the historical data includes the material quality, plate thickness of the plate transmitted from the rough rolling exit to the finishing rolling entry, and temperature change data under the corresponding cooling condition;

[0008] According to the real-time collected temperature data and the output result of the temperature prediction control model, the parameters of the temperature prediction control model are dynamically adjusted;

[0009] Based on the deviation between the temperature distribution predicted by the temperature prediction control model and the target temperature distribution, the parameters of the cooling system are adjusted to control the cooling temperature at the finishing rolling entry.

[0010] In some optional implementations, the input parameters of the temperature prediction control model include the rough rolling outlet temperature, the temperature learning coefficient, the time difference, the radiation ratio, the specific heat, and the specific gravity; the output parameter of the temperature prediction control model includes the finish rolling inlet temperature; and the temperature prediction control model predicts the temperature distribution of each plate in the current cooling process by calculating the average value of the finish rolling inlet temperature.

[0011] In some optional implementations, the method for obtaining the input parameters of the temperature prediction control model includes: determining the rough rolling outlet temperature of the plate based on the data collected by the infrared pyrometer; determining the temperature learning coefficient based on the environmental temperature, the plate temperature, and the stage coefficient of different cooling stages; determining the time difference based on the first timestamp at which the temperature collection point of the plate reaches the rough rolling outlet and the second timestamp at which the temperature collection point reaches the finish rolling inlet; determining the value of the radiation ratio corresponding to the plate in the current cooling stage by using a preset relationship model between the radiation ratio and the plate characteristics; determining the value of the specific heat corresponding to the plate in the current cooling stage according to a relationship curve between the specific heat and the temperature determined in advance through experiments; and determining the value of the specific gravity of the plate according to the material of the plate.

[0012] In some optional implementations, the parameters of the temperature prediction control model are dynamically adjusted according to the real-time collected temperature data and the output result of the temperature prediction control model, including: comparing the predicted finish rolling inlet temperature output by the temperature prediction control model with the actually measured finish rolling inlet temperature to calculate the temperature deviation therebetween; dynamically correcting the temperature learning coefficient by using a feedback control algorithm according to the size and the change trend of the temperature deviation; and recalibrating at least one of the environmental temperature, the radiation ratio, the specific heat, or the time difference in the input parameters when the temperature deviation exceeds a preset threshold.

[0013] In some optional implementations, the parameters of the cooling system are adjusted to control the cooling temperature at the finish rolling inlet based on the deviation between the temperature distribution predicted by the temperature prediction control model and a target temperature distribution, including: predicting the temperature distribution in the current cooling process by using the adjusted temperature prediction control model; comparing the prediction result of the temperature distribution with the set target temperature distribution; and determining the parameter adjustment strategy of the cooling system to control the cooling temperature at the finish rolling inlet according to the range in which the comparison result exceeds the standard temperature deviation.

[0014] In some optional implementations, the parameters of the cooling system include cooling air speed, cooling time, and spray flow rate; determining the parameter adjustment strategy of the cooling system includes: constructing a three-dimensional geometric model of the cooling system according to the structure of the roughing exit area; the cooling system includes nozzles arranged on the conveying path of the plate, and the nozzles are used to spray air-water atomization to the plate; based on the target temperature of the current cooling stage and the preset cooling parameters, the cooling process of the plate is simulated by using the three-dimensional geometric model; and the parameter adjustment strategy of the cooling system is optimized based on the simulation result.

[0015] In some optional implementations, optimizing the parameter adjustment strategy of the cooling system based on the simulation result includes: determining the influence of different nozzle layouts, spray speeds, and spray angles on the cooling uniformity and cooling efficiency of the plate based on the simulation result; and generating a cooling parameter adjustment strategy according to the influence analysis result, the cooling parameter adjustment strategy including: adjusting the distribution density and arrangement mode of the nozzles on the conveying path of the plate; and adjusting the spray speed, spray angle, and / or spray time of the nozzles.

[0016] In a second aspect, an embodiment of the present application provides a finishing mill entrance temperature prediction control device, which includes:

[0017] a model construction unit configured to construct a temperature prediction control model based on historical data in a hot rolling and cooling process of different plates, the historical data including material quality, plate thickness, and temperature change data under corresponding cooling conditions of the plate conveyed from the roughing mill exit to the finishing mill entrance;

[0018] a model optimization unit configured to dynamically adjust parameters of the temperature prediction control model according to real-time collected temperature data and output results of the temperature prediction control model;

[0019] a temperature control unit configured to adjust parameters of a cooling system to control the cooling temperature at the finishing mill entrance based on a deviation between a temperature distribution predicted by the temperature prediction control model and a target temperature distribution.

[0020] In a third aspect, an embodiment of the present application provides an electronic device including a memory and a processor, the memory storing a computer program executable on the processor, and the processor implements steps of the method of any one of the first aspect when executing the computer program.

[0021] In a fourth aspect, an embodiment of the present application provides a storage medium storing computer executable instructions, the computer executable instructions causing the processor to execute the method of any one of the first aspect when invoked and executed by the processor.

[0022] The application provides a finishing entry temperature prediction control method and device, an electronic device and a storage medium, which first constructs a temperature prediction control model based on historical data in a hot rolling cooling process of different plates, the historical data including material quality, plate thickness and temperature change data under corresponding cooling conditions of the plate transmitted from a rough rolling outlet to a finishing entry; then parameters of the temperature prediction control model are dynamically adjusted according to real-time collected temperature data and an output result of the temperature prediction control model; and finally, parameters of a cooling system are adjusted to control the cooling temperature of the finishing entry based on a deviation between a temperature distribution predicted by the temperature prediction control model and a target temperature distribution. The above method solves the technical problem of low cooling temperature control precision in the hot rolling production process in the prior art, and achieves the technical effects of improving the finishing entry temperature stability, cooling efficiency and product quality stability. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 A flowchart of a finishing entry temperature prediction control method provided by the embodiment of the present application is shown in the figure.

[0025] Figure 2 A structural diagram of a cooling system of a hot rolled plate provided by the embodiment of the present application is shown in the figure.

[0026] Figure 3 A structural diagram of a finishing entry temperature prediction control device provided by the embodiment of the present application is shown in the figure.

[0027] Figure 4 A structural diagram of an electronic device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the application provided in the accompanying drawings is not intended to limit the scope of the application claimed, but merely represents selected embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the application.

[0030] It should be noted that similar reference numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Some embodiments of the application are described in detail below in conjunction with the accompanying drawings. The embodiments described below and the features in the embodiments can be combined with each other without conflict.

[0031] In the hot rolling production process, there are problems of temperature fluctuation, response lag and low cooling efficiency in the hot rolling finishing rolling inlet air cooling temperature control. Although the traditional PID control can adjust the temperature to a certain extent, the control precision is limited and it is difficult to adapt to the complex and changeable production environment. Some technologies use fuzzy control or digital means, but still face challenges such as poor environmental adaptability, high system complexity and insufficient real-time data processing capability.

[0032] Based on this, the present scheme proposes a finishing rolling inlet temperature prediction control method, device, electronic equipment and storage medium to solve the technical problem of low cooling temperature control precision in the hot rolling production process in the prior art.

[0033] In order to facilitate the understanding of the present embodiment, first, a finishing rolling inlet temperature prediction control method disclosed by the present embodiment is introduced in detail, referring to the flowchart of a finishing rolling inlet temperature prediction control method shown in Figure 1 The method can be applied to a cooling system of a hot rolled plate shown in Figure 2 The method mainly includes the following steps S110 to S130:

[0034] S110: Construct a temperature prediction control model based on historical data in the hot rolling cooling process of different plates, the historical data including the material quality, plate thickness and temperature change data under the corresponding cooling condition of the plate transmitted from the rough rolling outlet to the finishing rolling inlet;

[0035] The temperature prediction control model can be used to predict the temperature distribution of different types of plates in the cooling process. In the present embodiment, the material quality of the plate usually refers to the main material of the hot rolled plate, which can be low carbon steel, high strength alloy steel, silicon steel, heat-resistant steel or high-temperature alloy, etc.

[0036] The cooling condition can refer to factors affecting the heat transfer rate and temperature distribution of the plate during the cooling process, which can mainly include external environmental conditions (such as ambient temperature) and process parameters (such as parameters of the cooling system).

[0037] The temperature change data can refer to different temperature values corresponding to changes in the temperature of the plate surface over time or under corresponding cooling conditions, which can be acquired in real time by temperature sensors (such as infrared pyrometers or thermocouples, etc.) under corresponding cooling conditions.

[0038] The historical data can be temperature data collected under different cooling conditions from long-term production processes, which can be used to establish a temperature prediction control model.

[0039] S120: dynamically adjusting the parameters of the temperature prediction control model according to the real-time collected temperature data and the output results of the temperature prediction control model;

[0040] The real-time collected temperature data can include the exit temperature of the rough rolling and the entry temperature of the finish rolling, which can be collected by temperature sensors arranged at the corresponding positions (exit of rough rolling and entry of finish rolling).

[0041] As a specific example, the input parameters of the temperature prediction control model can include the exit temperature of the rough rolling, the temperature learning coefficient, the time difference, the radiation ratio, the specific heat, and the specific gravity; and the output parameter of the temperature prediction control model can include the entry temperature of the finish rolling. That is, the temperature prediction control model can predict the temperature distribution of each plate in the current cooling process by calculating the average value of the entry temperature of the finish rolling.

[0042] In one embodiment, the dynamic adjustment of the parameters of the temperature prediction control model can be achieved by the following method: first, comparing the predicted entry temperature of the finish rolling output by the temperature prediction control model with the actually measured entry temperature of the finish rolling to calculate the temperature deviation therebetween; then, according to the size and change trend of the temperature deviation, using a feedback control algorithm to dynamically correct the temperature learning coefficient; when the temperature deviation exceeds a preset threshold, recalibrating at least one of the ambient temperature, the radiation ratio, the specific heat, or the time difference in the input parameters to obtain the temperature prediction control model with updated parameters.

[0043] S130: adjusting the parameters of the cooling system to control the cooling temperature at the entry of the finish rolling based on the deviation between the temperature distribution predicted by the temperature prediction control model and the target temperature distribution.

[0044] The target temperature distribution can refer to specific temperature values that each part of the plate is expected to reach at the entry of the finish rolling, which can be pre-set according to production process requirements and product quality standards.

[0045] In an embodiment, the parameters of the cooling system include cooling air speed, cooling time and spray flow rate, which can be used to correct the cooling rate and adjust the temperature distribution.

[0046] Further, the implementation of S130 can include predicting the temperature distribution in the current cooling process by using the adjusted temperature prediction control model; comparing the prediction result of the temperature distribution with the set target temperature distribution; and determining the parameter adjustment strategy of the cooling system according to the comparison result exceeding the standard temperature deviation range, so as to control the cooling temperature at the finishing mill inlet.

[0047] In an embodiment, the temperature prediction control model can be constructed based on a multiple regression model or a machine learning model. Further, the input parameters of the temperature prediction control model can include the rough rolling outlet temperature, the temperature learning coefficient, the time difference, the radiation ratio, the specific heat and the specific gravity; and the output parameter of the temperature prediction control model can include the finishing mill inlet temperature. That is, the temperature prediction control model can predict the temperature distribution of each plate in the current cooling process by calculating the average value of the finishing mill inlet temperature (i.e. the temperature learning effect).

[0048] As a specific example, the formula for calculating the average value of the finishing mill inlet temperature is as follows:

[0049] (Formula 1);

[0050] wherein, Fet Cal represents the average value of the finishing mill inlet temperature, in ℃; Rdt Cal represents the average value of the rough rolling outlet temperature, in ℃; Ep Con represents the radiation ratio; Sig Con represents the Stefan-Boltzmann constant, in kcal / m 2 hr℃ 4 ; Tim(1) represents the time difference of the plate from the rough rolling outlet to the finishing mill inlet, in hr; Cp Con represents the specific heat t, in kcal / kg℃; Gam Con represents the specific gravity, in kg / m 3 ; Lct Lay represents the temperature learning coefficient; Rdh Mod represents the rough rolling outlet thickness, in mm.

[0051] In the present embodiment, the average value of the above-mentioned finish rolling inlet temperature can be used to dynamically adjust the key control parameters of the hot-rolled plate in the finish rolling inlet cooling process, so as to realize accurate prediction of the temperature distribution and control of the cooling rate, thereby optimizing the cooling process and improving the overall performance and quality stability of the plate; at the same time, through the calculation of the average value of the temperature learning effect, the temperature variation law in the cooling process can be quantified, thereby providing a scientific basis for subsequent process adjustment and parameter optimization.

[0052] Further, gradient descent or other optimization algorithms can also be used to gradually optimize the above-mentioned temperature prediction control model according to the deviation of real-time temperature data from the target temperature distribution. As a specific example, the value of the temperature learning coefficient can be adjusted according to the error between the model-predicted temperature distribution and the actual distribution, so as to satisfy the optimization condition:

[0053] (Formula 2);

[0054] wherein, is the learning rate, and E is the prediction error.

[0055] After updating the parameters, the optimized temperature learning coefficient is substituted into the formula to recalculate the average value of the finish rolling inlet temperature (i.e., the temperature learning effect).

[0056] Further, in some embodiments, the input parameters of the above-mentioned temperature prediction control model can be obtained in the following manner:

[0057] (1) The initial rolling outlet temperature of the plate is determined by the data collected by the infrared pyrometer.

[0058] The infrared pyrometer can be a non-contact infrared pyrometer installed at the initial rolling outlet, which calculates the temperature value by collecting the infrared light radiated from the surface of the temperature collection point of the plate.

[0059] In order to ensure uniform temperature distribution of the entire plate, several key points (or key regions) on the plate can be selected as temperature collection points (or temperature collection regions). For example, the geometric center point of the plate can be selected as the temperature collection point to collect the temperature of the geometric center of the plate; or a point at the head of the plate can be selected as the temperature collection point to collect the temperature of the plate just after leaving the initial rolling outlet; or a point at the tail of the plate can be selected as the temperature collection point to collect the temperature of the plate after completely leaving the initial rolling outlet, etc.

[0060] In some other examples, multiple key points on the plate can also be selected as temperature collection points. By detecting the temperatures of these key points, the temperature distribution of the plate can be more comprehensively understood, so that corresponding control measures can be taken to ensure the stability of the production process and the product quality.

[0061] (2) determining the temperature learning coefficient based on the environment temperature, the plate temperature and the stage coefficient of different cooling stages;

[0062] In the process of hot rolling and cooling, the plate is usually transmitted from the initial rolling outlet to the finishing rolling inlet, and different cooling stages are experienced, and the heat transfer environment and cooling rate of different stages may be significantly different. In order to obtain more accurate prediction results, in some embodiments, the temperature learning coefficient can be adjusted according to different cooling stages.

[0063] As a specific example, the cooling stage can be divided into an initial rapid cooling stage, an intermediate stable cooling stage and a final slow cooling stage, and the temperature learning coefficient can be dynamically adjusted according to the current cooling stage and the heat transfer environment. Further, in the initial rapid cooling stage, since the temperature gradient is large and the heat transfer rate is high, the temperature learning coefficient should be large to reflect the rapid temperature change; in the intermediate stable cooling stage, since the temperature gradient gradually decreases and the heat transfer rate also decreases accordingly, the temperature learning coefficient should be appropriately reduced; in the final slow cooling stage, since the temperature gradient is very small and the heat transfer rate is very low, the temperature learning coefficient should be further reduced.

[0064] For example: the temperature learning coefficient of the initial rapid cooling stage alpha = alpha0 + k1 x (Tsteel - Tenv), the temperature learning coefficient of the intermediate stable cooling stage alpha = alpha0 + k2 x (Tsteel - Tenv), and the temperature learning coefficient of the final slow cooling stage alpha = alpha0 + k3 x (Tsteel - Tenv); wherein Tsteel is the current temperature of the plate, Tenv is the current environment temperature; the basic learning coefficient alpha0 = 0.5, the initial stage proportion constant k1 = 0.1, the stable stage proportion constant k2 = 0.05, and the final stage proportion constant k3 = 0.01. In actual application, these parameters can be verified and adjusted through experimental data to ensure that the temperature learning coefficient can accurately reflect the heat transfer environment of different cooling stages.

[0065] In this embodiment, through dynamic adjustment of the temperature learning coefficient, the control system for executing the finishing rolling inlet temperature prediction control method provided by the embodiment of the application can flexibly adjust the parameters according to the actual characteristics of the plate.

[0066] As a specific example, the temperature learning coefficient Lct LayThe dynamic adjustment can be achieved in the following way: first, by collecting the surface temperature and internal temperature of the plate during the cooling process in real time, combining historical data (including cooling laws under different material, thickness and process conditions), a data-driven temperature learning coefficient model is established.

[0067] Then, using the temperature learning coefficient model, the deviation of real-time temperature change from the target temperature is fitted and predicted; the key influencing factors (such as initial temperature, time difference, plate thickness and cooling wind speed) in the cooling process are extracted, and the weight parameters of the temperature learning coefficient model are dynamically updated. According to the prediction result of the temperature learning model, the value of the temperature learning coefficient Lct Lay is adjusted to the optimal value suitable for the current cooling environment.

[0068] (3) Determine the time difference by the first timestamp of the temperature collection point of the plate reaching the rough rolling outlet and the second timestamp of reaching the finish rolling inlet;

[0069] Among them, the first timestamp can be the first moment recorded by the synchronous clock system when the temperature collection point of the plate reaches the rough rolling outlet; the second timestamp can be the second moment recorded by the synchronous clock system when the temperature collection point of the plate reaches the finish rolling inlet, and the time difference is the difference between the second moment and the first moment.

[0070] As a specific example, the way to determine the time difference can include: collecting the temperature data of the plate (at the temperature collection point) at the rough rolling outlet by the non-contact infrared pyrometer installed at the rough rolling outlet (the accurate temperature value is calculated after the infrared light radiated by the plate surface is converted into an electrical signal), and recording the time point t1 passing through the rough rolling outlet; detecting the surface temperature distribution and real-time change of the plate by the non-contact infrared pyrometer arranged at the finish rolling inlet, and detecting the position change of the plate (at the temperature collection point) reaching the finish rolling inlet by the position sensor, and recording the time point t2 reaching; and the time difference .

[0071] (4) Determine the value of the radiation ratio corresponding to the current cooling stage of the plate by using the pre-set relationship model between the radiation ratio and the characteristics of the plate;

[0072] The radiation ratio refers to the ratio of the ability of an object's surface to emit thermal radiation to the emission ability of an ideal black body, and is usually used for thermal radiation calculation. In the present scheme, the radiation ratio is used to calculate the radiation heat transfer of the hot-rolled plate during the cooling process, and the Stefan-Boltzman law is used to estimate the thermal radiation loss of the hot-rolled plate during the cooling process, and the radiation ratio is a key parameter.

[0073] Radiation heat transfer is one of the main ways of heat dissipation for plate under high temperature conditions, so the accurate radiation ratio value is crucial for the accuracy of temperature prediction. Different materials (such as different steel grades) have different radiation ratios. Through experiments, accurate radiation ratio values can be determined to improve the accuracy of heat loss calculation during cooling, thereby improving the accuracy of the prediction of the finishing mill inlet temperature.

[0074] The method for obtaining the relationship model between the radiation ratio and the plate characteristics can include: repeatedly performing experiments to obtain radiation ratio data of a plurality of plate samples under different temperatures and surface conditions, and then performing regression analysis on the experimental results to fit the relationship model between the radiation ratio and the plate characteristics.

[0075] Further, in the relationship model between the radiation ratio and the plate characteristics, the fitted radiation ratio formula or the table lookup method is used to select the applicable radiation ratio value according to the material and cooling conditions of the plate, so as to determine the value of the radiation ratio corresponding to the pre-cooling stage.

[0076] In some embodiments, the above-mentioned experiment for obtaining radiation ratio data can be implemented by using a radiation ratio experimental device to perform the following process. Specifically, the radiation ratio experimental device can include: a radiation heat flow measuring device (composed of a high-precision heat flow meter and an infrared radiation thermometer, which are respectively used to measure the heat flux density q of the plate surface radiation and the real-time temperature T of the plate surface S ), a temperature control experimental furnace (used to simulate the cooling environment of the hot-rolled plate, to adjust the environmental temperature T e and the radiation intensity), and a plate sample (which can be selected from a plurality of different materials including but not limited to low-carbon steel, high-strength alloy steel, silicon steel, heat-resistant steel, and high-temperature alloy, and has a sample with an untreated original hot-rolled surface and a thick surface oxide layer). In this embodiment, the material of the plate sample is selected to be low-carbon steel and high-strength alloy steel, and the thickness range is 2 mm, 4 mm, and 6 mm.

[0077] As a specific example, the process of the radiation ratio experiment includes:

[0078] Step 1, sample preparation: clean the surface of the plate sample, and record the material, thickness, surface roughness, and color characteristics thereof;

[0079] Step 2, experimental setup: place the sample in the experimental furnace and heat it to a target temperature close to the actual initial rolling outlet temperature;

[0080] Step 3, measuring heat flow and temperature: use an infrared radiation thermometer (or a high-precision infrared thermometer) to detect the surface temperature of the sample in real time; use a heat flow meter to measure the heat flux density of the radiation from the sample surface to the outside;

[0081] Step 4, calculating the radiation ratio: calculate the radiation ratio according to the Stefan-Boltzmann law, and the calculation formula is:

[0082] (Equation 3);

[0083] where q represents the heat flux density of the sample surface radiation (unit: W / m 2 ); Sig Con represents the Stefan-Boltzmann constant (5.67 x 10 ); T S represents the sample surface temperature (unit: K); T e represents the ambient temperature (unit: K).

[0084] Step 5, data processing: repeat the experiment to obtain the radiation ratio data of multiple samples under different temperatures and surface conditions; perform regression analysis on the experimental results to fit the relationship model between the radiation ratio and the sample characteristics;

[0085] Step 6, result application: use the fitted radiation ratio formula or table lookup method in the model to select the applicable radiation ratio value according to the plate material and cooling conditions.

[0086] Further, the value of the radiation ratio Ep Con can be changed through experiments to study its influence on the cooling performance of the plate (including the cooling rate and temperature distribution uniformity) to determine the radiation ratio value that can achieve the best cooling effect. In one embodiment, the experimental method for determining the optimal radiation ratio and data analysis can include the following steps:

[0087] <1>, the experimental device includes:

[0088] 1. A high-precision heat flow meter for measuring the heat flux density q of the plate surface radiation;

[0089] 2. An infrared radiation thermometer for measuring the real-time temperature T s of the plate surface;

[0090] 3. A temperature-controlled experimental furnace for simulating hot rolling environment and adjusting the ambient temperature T e and radiation intensity;

[0091] The plate samples are selected to be low-carbon steel and high-strength alloy steel; the thickness range of the plate samples is 2 mm, 4 mm, and 6 mm.

[0092] <2>, the experimental steps are as follows:

[0093] 1. Sample preparation: surface treatment (polishing, spraying oxidation layer, etc.) is performed on the plate samples to change their radiation ratio values; samples with different surface states (radiation ratio range Ep Con = 0.3-1.1) are used.

[0094] 2. Experimental setup: heat the sample to 900-1200°C; set the ambient temperature Te = 50 °C; control sample surface emissivity Ep Con , record their corresponding cooling rate and final temperature distribution.

[0095] 3, data acquisition: real-time record sample surface temperature T s (t), ambient temperature T e , and the radiant heat flux q.

[0096] 4, data processing: using the Stefan-Boltzmann formula to calculate the heat transfer efficiency under different radiation ratio:

[0097] (Equation 4);

[0098] Where q is the radiation heat transfer heat flux (i.e. sample surface radiation heat flux), represents the heat radiation transfer per unit area (W / m 2 ); Ep Con is the sample surface emissivity (i.e. radiation ratio); Sig Con is the Stefan-Boltzmann constant ( ); T s is the sample surface temperature (K); T e is the ambient temperature (K).

[0099] The experimental results are shown in Table 1 as follows:

[0100] Table 1 - Radiation ratio experimental results;

[0101]

[0102] <3>, analysis of experimental results:

[0103] 1, the relationship between cooling rate and radiation ratio: when the radiation ratio Ep Con increases, the heat flux q increases, and the cooling rate R c increases accordingly. When Ep Con > 0.7, the cooling rate growth tends to be flat.

[0104] 2, temperature distribution uniformity: when the radiation ratio is 0.7, the temperature uniformity is best (deviation is the smallest). When the radiation ratio is too large (such as Ep Con = 1.1), the surface supercooling phenomenon increases, leading to the expansion of the temperature difference between the inside and the surface.

[0105] 3, determination of the optimal radiation ratio: considering the cooling rate and temperature uniformity, the optimal radiation ratio is 0.9.

[0106] Further, the specific heat and the specific gravity in the input parameters of the temperature prediction control model can be determined by experiments, which are used to accurately calculate the energy transfer and temperature change of the hot-rolled plate in the cooling process at the entrance of the finishing mill, so as to improve the control accuracy of the cooling process; the specific heat and the specific gravity determined by experiments can provide reliable data support for the calculation of the average value of the temperature learning effect, so as to optimize the heat transfer model in the cooling process, improve the cooling efficiency, and ensure the uniformity of the plate temperature distribution and the consistency of the product performance.

[0107] (5) According to the relationship curve between the specific heat and the temperature determined by experiments in advance, the specific heat value corresponding to the plate in the current cooling stage is determined;

[0108] The specific heat refers to the heat absorbed by unit mass of a substance to raise the temperature by one unit, and the unit is generally kcal / (kg·℃). As an inherent thermal physical property of a material, the specific heat can be used to characterize the response ability of the material to temperature change. In the cooling process, the specific heat determines how much heat is absorbed or released by the plate when the temperature changes.

[0109] In the present scheme, the specific heat can be used to calculate the heat released by the plate in the cooling process, which, together with the temperature change and the mass, determines the energy change of the plate in the cooling process, and is an important parameter for calculating the temperature change of the plate in the cooling process. By determining the specific heat values of different materials through experiments, the model can be more consistent with the actual working conditions, and the reliability of temperature prediction can be improved.

[0110] In an example, the way of determining the specific heat value by experiments can include: testing the main material (such as low-carbon steel or high-strength alloy steel) of the hot-rolled plate by a differential scanning calorimeter (DSC), placing a standard size sample in a temperature-controlled environment, gradually heating from room temperature to a target temperature, recording the heat absorbed by the sample, the mass and the temperature change, and calculating the specific heat value according to the following formula:

[0111] (Formula 5);

[0112] Wherein, Q represents the heat absorbed (J), m represents the mass of the sample (kg), and △t represents the temperature difference;

[0113] According to the experimental results, a relationship curve between the specific heat and the temperature is generated, which provides accurate thermal physical parameter support for the calculation of the temperature learning effect in the model.

[0114] In an embodiment, the specific method and data analysis of experimentally determining the specific heat can include the following steps:

[0115] <1> The experimental method is differential scanning calorimetry (DSC); the experimental device includes:

[0116] 1. Differential Scanning Calorimeter (DSC): Used to measure the relationship between the heat absorbed or released by a sample and the temperature.

[0117] 2. Temperature control environment: Temperature control accuracy is ±0.1℃, and the test temperature range is 25~1000℃.

[0118] 3. Standard reference material: Use materials with standard specific heat value (such as alumina) for instrument calibration.

[0119] II. Experimental Procedure:

[0120] 1. Sample preparation: Take a standard sample of low-carbon steel or high-strength alloy steel, with a mass of 0.5g; clean the sample surface to ensure that there are no impurities or oxide scale.

[0121] 2. Experimental setup: Place the sample and reference material in the sample pan and reference pan of the DSC instrument, respectively; set the heating rate to 10K / min and the target temperature range to 25–900℃.

[0122] 3. Data acquisition: Record the heat Q absorbed by the sample at different temperatures, the corresponding temperature change ΔT, and the sample mass m.

[0123] 4. Specific heat calculation formula:

[0124] (Formula 5);

[0125] Among them, Cp Con Δt represents specific heat (J / kg•℃); Q represents the heat absorbed by the sample (J); m represents the sample mass (kg); and Δt represents the temperature change (℃). The experimental data are shown in Table 2 below.

[0126] Table 2 - Experimental test results of specific heat;

[0127]

[0128] <III> Analysis of Experimental Results: The specific heat value of low-carbon steel increases with increasing temperature, increasing from 2000 J / kg•℃ to 3200 J / kg•℃ within the range of 25–400℃. The specific heat value of high-strength alloy steel shows a similar trend, but the absolute value is slightly higher.

[0129] (6) Determine the specific gravity of the board based on its material.

[0130] Specific gravity refers to the mass of a substance per unit volume, and its unit is kg / m³. 3As one of the basic physical properties of materials, the specific gravity determines the heat capacity of the material in a unit volume. In the present scheme, the specific gravity is used to calculate the mass of the unit volume of the plate, and further used to calculate the energy change. In combination with parameters such as specific heat and thickness, the specific gravity determines the overall heat capacity of the plate. The greater the mass of the material, the more heat it contains during the cooling process, and the slower the cooling rate. The specific gravity participates in the calculation of the overall heat energy change and is a key factor affecting the temperature change rate.

[0131] For different types of hot-rolled plates (such as different steel grades), the specific gravity may be different. By experimentally determining the value of the specific gravity, the accuracy of the energy transfer calculation in the model can be improved, and the adaptability of the model to different materials can be further improved.

[0132] As a specific example, the way of experimentally determining the specific gravity value corresponding to different plate materials can include: testing the main material of the hot-rolled plate (such as low-carbon steel or high-strength alloy steel) by Archimedes method, measuring the sample mass m by high-precision electronic scale, then completely immersing the sample in a liquid of known density (such as water), and measuring the volume V of the liquid displaced; calculating the specific gravity value according to the ratio of the sample mass m to the volume V of the liquid displaced, and taking the average value of multiple sample tests as the specific gravity value corresponding to the plate material of the current sample.

[0133] In one embodiment, the specific method of experimentally determining the specific gravity and data analysis can include the following steps:

[0134] <1>, the experimental method is Archimedes method; the experimental device includes:

[0135] electronic balance (accuracy 0.001 g); liquid of known density, pure water (density 1000 kg / m 3 ); experimental fixture for fixing the sample to prevent movement during measurement.

[0136] <2>, experimental steps:

[0137] 1, sample preparation: select low-carbon steel or high-strength alloy steel sample, mass 5g; clean the sample surface to ensure no attachments.

[0138] 2, experimental setup: measure the mass of the sample in air m dry ; completely immerse the sample in pure water and measure the buoyancy mass after displacing water m buoyancy .

[0139] 3, specific gravity calculation formula:

[0140] sample volume V sample , as follows:

[0141] (Formula 6);

[0142] Specific gravity , as follows:

[0143] (Formula 7);

[0144] The experimental data are shown in Table 3 as follows:

[0145] Table 3 - Experimental test results of specific gravity

[0146]

[0147] <3>, analysis of experimental results: the specific gravity value of low carbon steel is about 7850kg / m 3 ; the specific gravity value of high-strength alloy steel is slightly higher, about 8900kg / m 3 .

[0148] The above experimental results can provide accurate density parameter support for the calculation of temperature learning effect, to ensure the reliability and accuracy of model calculation.

[0149] Taking the experimental test results in the above examples as an example, combined with the real-time acquisition of model parameters, the process of calculating the average value of the finish rolling inlet temperature can include the following steps:

[0150] Real-time acquisition of model parameters, including: the average temperature Rdt Cal at the rough rolling outlet, which is 1127°C in real-time detection; the radiation ratio Ep Con , which is determined by experiment to be 0.35; the Stefan-Boltzmann constant Sig Con , ; the time difference Tim(1) from the rough rolling outlet to the finish rolling inlet, which is 0.004hr; the specific heat Cp Con : experimentally determined to be 0.9kcal / kg∙℃; the specific gravity Gam Con : experimentally determined to be 8850kg / m 3 ; the temperature learning coefficient Lct Lay , the initial value is 1.0, which is optimized to 1.02 after dynamic adjustment; the rough rolling outlet thickness Rdh Mod , which is 0.065m in real-time monitoring.

[0151] The above parameters are brought into the formula (i.e. Formula 1) of the average value of the finish rolling inlet temperature for calculation, and the following results are obtained:

[0152] .

[0153] The present scheme determines the physical parameters such as radiation ratio, specific heat and specific gravity through experiments, avoids the errors caused by relying on empirical values or fixed values, and thus improves the prediction accuracy and control stability of the temperature prediction control model.

[0154] In another embodiment, the S120 dynamically adjusts the parameters of the temperature prediction control model according to the real-time collected temperature data and the output results of the temperature prediction control model, which can include:

[0155] (S121) comparing the predicted finish rolling inlet temperature output by the temperature prediction control model with the actually measured finish rolling inlet temperature, and calculating the temperature deviation therebetween; (S122) dynamically correcting the temperature learning coefficient by using a feedback control algorithm according to the size and variation trend of the temperature deviation; (S123) when the temperature deviation exceeds a preset threshold, recalibrating at least one of the environmental temperature, the radiation ratio, the specific heat or the time difference in the input parameters.

[0156] Through the dynamic correction of the model parameters as described above, the error between the predicted value and the actually measured value can be effectively reduced, and the prediction accuracy of the model can be significantly improved; the feedback control algorithm is used to dynamically correct the temperature learning coefficient, so that the model can be self-adjusted according to the variation trend of the temperature deviation, thereby enhancing the adaptability and robustness of the model under different working conditions, and being applicable to the continuous production process under different plate materials and different environmental conditions; through the recalibration of the input parameters, the model failure caused by the drift of the input parameters or the measurement error can be prevented, and the fault tolerance and stability are improved.

[0157] (S124) based on the historical deviation data generated by the plurality of plates during the cooling process, using the least square method or the Kalman filtering method to perform online adaptive optimization and update of the model parameters; through this step, the model can have the ability of continuous learning and optimization, and can adapt to the parameter drift and process changes in long-term operation.

[0158] (S125) after updating the model parameters, the finish rolling inlet temperature prediction value is recalculated, and the updated model parameters are stored in the control system for temperature prediction and cooling control of subsequent plates. Through this step, a closed-loop optimization control process can be formed, and complete closed-loop management from data acquisition, model prediction, deviation analysis, parameter updating to control execution can be realized.

[0159] Based on this, the embodiment introduces a dynamic feedback mechanism, an abnormal processing mechanism and an online parameter optimization mechanism, and constructs a temperature prediction control model with high precision, self-adaptation and continuous learning ability. Through the model, more accurate temperature control can be realized, which helps to improve the cooling uniformity and organizational performance of the plate, reduces the quality defects caused by poor temperature control, and improves the stability and efficiency of the production process.

[0160] In one embodiment, the parameters of the cooling system can include: cooling air speed, cooling time and jet flow; further, determining the parameter adjustment strategy of the cooling system can include:

[0161] (S131) ​​Based on the structure of the finishing mill inlet region, construct a three-dimensional geometric model of the cooling system;

[0162] The finishing mill entrance area may include a cooling system and a finishing mill stand F1, combined with... Figure 2 As shown, the cooling system may include: nozzle 210, guide 220, roll 230, control module 240 and temperature sensor 250.

[0163] Among them, nozzle 210 is set at the exit of the primary rolling mill (where the plate exits from the primary rolling mill). Figure 2 A) is transferred to the finishing mill inlet ( Figure 2 In the transmission path of (B) in the diagram, the nozzle can be located above, below, or to the side of the board. It is used to spray air-water atomization (an atomized cooling medium formed by a mixture of compressed air and cooling water) onto the board to remove impurities from the board surface and cool the board. The position of the nozzle, the spray angle, the spray pressure, and the coverage area of ​​the spray area can all affect the parameters of the cooling system, thereby affecting the temperature of the board surface.

[0164] The guide 220 can be installed on the finishing mill inlet side, located on both sides or above and below the plate's travel path, to guide the plate accurately into the gap of the rolls 230, prevent deviation, and support the plate's stable operation before entering the finishing mill inlet. The rolls 230 can be installed on the finishing mill stand F1, serving as the core execution component for the first pass of finishing milling, typically consisting of two rolls forming the rolling gap for the plate to pass through.

[0165] The control module 240 is connected to the nozzle 210 (electrical or communication connection) to control the output of the nozzle 210, such as controlling the number of nozzles 210 that are opened (to adjust the distribution density and arrangement of the nozzles on the transmission path of the board); adjusting the spray speed, spray angle and / or spray time of the nozzles, etc.

[0166] Temperature sensors 250 can be installed at the primary rolling exit A and the finishing rolling inlet B respectively to collect the temperature of the plate at the primary rolling exit and the finishing rolling inlet in real time. Temperature sensors 250 can also be used to collect the temperature of the surrounding environment.

[0167] Furthermore, the temperature sensor 250 can be connected to the control module 240 (electrical or communication connection) to transmit the collected temperature signal to the control module 240. The control module 240 can dynamically adjust the parameters of the cooling system (such as the layout of the nozzle 210, the injection speed, the injection angle, etc.) according to the target temperature and preset cooling parameters to achieve closed-loop control of the cooling process, thereby improving cooling uniformity and process stability.

[0168] (S132) Based on the target temperature of the current cooling stage and the preset cooling parameters, the cooling process of the plate is simulated by using the three-dimensional geometric model; and the parameter adjustment strategy of the cooling system is optimized based on the simulation results.

[0169] In an embodiment, optimizing the parameter adjustment strategy of the cooling system based on the simulation results can include: determining the influence of different nozzle layouts, jet wind speeds and jet angles on the cooling uniformity and cooling efficiency of the plate based on the simulation results; and generating a cooling parameter adjustment strategy according to the influence analysis results. The cooling parameter adjustment strategy can include: adjusting the distribution density and arrangement mode of the nozzles on the conveying path of the plate; adjusting the jet speed, jet angle and / or jet time of the nozzles to control the cooling wind speed and cooling time to match the heat conduction requirement of the current cooling stage.

[0170] In addition, in some examples, the cooling parameter adjustment strategy can also include adjusting the jet angle of the nozzles to improve the coverage of the cooling medium on the surface of the plate.

[0171] Further, the parameter adjustment strategy can be fed back to the control module of the cooling system to realize closed-loop optimization control of the cooling parameters.

[0172] In the embodiment, by combining the Computational Fluid Dynamics (CFD) technology, the fluid flow and heat conduction of the air cooling process of the hot-rolled plate at the entry area of the finishing mill are jointly modeled and simulated to improve the simulation accuracy and control ability of the cooling process.

[0173] As a specific example, the above method can include the following steps:

[0174] (S141) Establishing a three-dimensional geometric model of the cooling zone: according to the actual structure of the entry area of the finishing mill, a refined three-dimensional model containing the plate, nozzles, guides and rollers, etc. is constructed.

[0175] (S142) Setting boundary conditions and moving walls: the plate is set as a moving wall to consider the disturbance of its movement to the airflow field; the air inlet velocity, temperature and turbulence parameters are set, the outlet adopts a pressure boundary condition, and the wall surface sets a heat conduction and convection heat transfer boundary.

[0176] (S143) Selecting suitable turbulence and radiation models: selecting a suitable turbulence model such as k-ε or k-ω to simulate the flow characteristics of the spraying air, and selecting a suitable radiation model (P1, DO, etc.) to perform coupled calculation on the radiation heat transfer of the high-temperature surface of the plate.

[0177] (S144) Coupling convection and conduction mechanisms: Automatically calculate the local heat transfer coefficient through the convection heat exchange module of the CFD simulation software, while solving the internal heat conduction process of the plate, to realize the synchronous evolution of the temperature field inside and outside the plate.

[0178] (S145) Nozzle layout and cooling parameter optimization: Use the simulation model to simulate and compare multiple schemes of nozzle arrangement, wind speed, spray angle, etc., to evaluate the uniformity and efficiency of the cooling effect, and optimize the cooling system design.

[0179] (S146) Build an offline prediction model and support closed-loop control: Based on a large number of CFD simulation results, build a response surface model, proxy model or neural network model to realize fast prediction of the relationship between cooling parameters and temperature evolution. In actual production, combined with rolling speed, plate thickness, initial temperature and other parameters, realize closed-loop adjustment control of the cooling process.

[0180] Through the above implementation, the prediction accuracy, adaptability and intelligent level of the cooling system can be significantly improved, especially for hot-rolled high-strength steel, silicon steel and other product scenarios with high cooling uniformity and temperature control requirements.

[0181] As a specific example, to realize accurate simulation of the plate cooling process, first build a three-dimensional geometric model including the plate, cooling system and environment, and set the simulation conditions for multi-physics field coupling. Preferably, geometric modeling can be established using CAD software (such as SolidWorks, ANSYS DesignModeler) to establish a three-dimensional geometric model of the plate (size L x W x H, where L L is the length, W W is the width, H H is the thickness), nozzle array (candidate layout contains N nozzles, coordinates are x i , y i , z i ), i =1,2,…, N ) and transmission path (speed vtrans ).

[0182] Plate material properties: density ρ s (kg / m 3 ), specific heat capacity c p,s (J / (kg·K)), thermal conductivity k s (W / (m·K)), which can be obtained through material manual or differential scanning calorimetry (DSC) experiment.

[0183] Cooling medium properties: density of air (or water) ρ f(kg / m 3 ), dynamic viscosity μ (Pa·s), thermal conductivity k f (W / (m·K)), specific heat capacity at constant pressure c p,f (J / (kg·K)), the ambient temperature T ∞ can be measured by a temperature sensor and then looked up in a table.

[0184] Boundary conditions setting: initial condition: the initial temperature T0 (℃) of the plate is measured by an infrared thermal imager; surface condition: the upper surface of the plate is a spray cooling boundary, and the convective heat transfer and radiative heat transfer loads are applied; the lower surface is in natural convective heat transfer with the environment (the surface heat transfer coefficient hnat is taken as 5-10 W / (m 2 ·K)); nozzle outlet condition: the spray medium temperature T j (℃, set by a temperature control system) and the speed v j (m / s, to be optimized variable).

[0185] The simulation tool can use computational fluid dynamics (CFD) software (such as ANSYS Fluent, COMSOL Multiphysics) for multi-physical field coupling simulation, the solver can be set to a pressure-based transient solver, the turbulent flow model can use the Realizable k-ε model (suitable for high Reynolds number flow), and the radiation model can use the DO model (considering the influence of thermal radiation on the cooling process).

[0186] Through simulation of different cooling parameter combinations, the influence of nozzle layout, spray wind speed and spray angle on the cooling uniformity and cooling efficiency of the plate can be quantitatively analyzed. For example:

[0187] (1) Cooling uniformity evaluation index:

[0188] The cooling uniformity is measured by the standard deviation σT of the plate surface temperature distribution, and the calculation formula is:

[0189] (Formula 8);

[0190] In the formula: T(x, y) is the temperature (℃) at the (x, y) position of the plate surface;

[0191] is the average temperature (℃) of the surface; A = L × W is the area (m 2 ) of the upper surface of the plate. The smaller σT is, the more uniform the temperature distribution is.

[0192] (2) Cooling efficiency evaluation index:

[0193] The cooling efficiency η is defined as the ratio of the actual heat transfer quantity Qactual to the ideal heat transfer quantity Qideal, and the calculation formula is:

[0194] Equation 9

[0195] Qactual = Qideal x (1 - η) actual Qactual can be obtained by integrating the heat flux from simulation:

[0196]

[0197] Qideal = Qmax x η ideal Qmax is the theoretical maximum heat transfer when the plate is cooled from initial temperature T0 to ambient temperature T ∞

[0198] h eff is the effective convective heat transfer coefficient (W / (m 2 ·K), which is calculated from the Nusselt number (Nu) from simulation:

[0199] ε is the emissivity of the plate surface (dimensionless), which can be measured by a spectrometer;

[0200] σ= is the Stefan-Boltzmann constant; A s is the surface area of the plate in contact with the cooling medium (m 2 ); V = L x W x H is the volume of the plate (m 3 ).

[0201] The larger η is, the higher the cooling efficiency is.

[0202] Further, based on the simulation results, an optimization algorithm (such as genetic algorithm, particle swarm optimization) can be used to establish the mapping relationship between the cooling parameters and the evaluation index (η σT , η ), and to generate the optimal parameter adjustment strategy. One preferred way is:

[0203] (1) Nozzle layout optimization;

[0204] The optimization variables can include: nozzle distribution density ρ d (m 2 ) and arrangement mode (such as rectangular array, circular array, gradient array).

[0205] If the distribution density ρ d is too small, there will be uncovered areas on the plate surface, leading to local temperature being too high; if ρ d is too large, the mutual interference between nozzles will increase, and the energy utilization rate will decrease. Moreover, the gradient array (higher density in the center than at the edges) can compensate for the insufficient cooling at the edges due to the weakening of convective heat transfer (simulation shows that the temperature difference between the center and the edges can be reduced by 15-20%).

[0206] ​​​​Therefore, the optimization method can include: searching for the optimal p by a genetic algorithm with the objective of minimizing σT d And the arrangement, the fitness function is defined as:

[0207] (Formula 10);

[0208] In the formula, ω1, ω2 are weight coefficients (ω1+ω2=1); C nozzle is the cost of a single nozzle (yuan / each), which can be used to balance uniformity and economy.

[0209] (2) Optimization of jet wind speed and angle;

[0210] The optimization variables include: jet wind speed v j (m / s), jet angle α (°, the angle with the normal of the plate surface).

[0211] When the wind speed v j increases, the convective heat transfer coefficient h eff increases (h eff ∝v j ), and the cooling rate increases, but the energy consumption (fan power ) increases significantly; when the angle α is 45°, the projection area of the cooling medium on the plate surface is the largest (coverage C cover =cosα+sinα, when α=45°, C cover =1.414), and the rebound loss can be reduced (simulation shows that the surface coverage of α=45° is increased by 20% compared with vertical jet (α=90°)).

[0212] Therefore, the optimization method can include: searching for the optimal p by a genetic algorithm with the objective of minimizing σT j and maximizing η, and determining the optimal (v c , α) combination by Pareto frontier analysis, and the specific formula is:

[0213] (Formula 11);

[0214] (3) Jet time adjustment;

[0215] Optimization variable: single-stage jet time t c (s).

[0216] Because too long t c will cause the plate to be supercooled (the temperature is lower than the target value), and too short t target cannot fully dissipate heat.

[0217] Therefore, the optimization can be based on a heat conduction demand matching model to calculate the minimum t c that satisfies the target temperature T req (t).

[0218] (Equation 12);

[0219] wherein: q req (t) is the target heat flux (W / m 2 ), expressed as the heat flux required to meet T target ( t ) is the instantaneous heat exchange between the cooling medium and the plate; p s is the plate density (unit: kg / m 3 );

[0220] δ ( t ) is the dynamic thermal penetration depth (m), used to represent the effective depth of temperature variation impact, with an initial value ( is the plate thermal diffusivity, t 0 is the initial cooling time), and the final form is ( kδ is the adjustment coefficient, τ is the time constant, both of which are calibrated through simulation or experiment); is the rate of change of the target temperature over time (unit: K / s), which can be obtained by differentiating the target temperature curve T target ( t ) (determined through process planning or experimental measurement).

[0221] Next, the optimized cooling parameters (nozzle layout, v j , α , t c ) can be transmitted to the PLC (Programmable Logic Controller) of the cooling system through industrial Ethernet (such as PROFINET), to real-time adjust the nozzle driving mechanism (such as step motor control nozzle angle), frequency converter (adjust fan speed to change v j ), and electromagnetic valve (control injection time t c ). At the same time, the plate surface temperature is monitored in real-time through an infrared thermal imager, forming a closed-loop control of "simulation optimization-parameter adjustment-real-time monitoring-reoptimization", to ensure that the cooling process is always in the optimal state.

[0222] This embodiment realizes precise adjustment of cooling parameters through three-dimensional simulation and multi-objective optimization, solves the problem of difficult trade-off between uniformity and efficiency in traditional experience adjustment, and provides technical support for intelligent control of the plate cooling process.

[0223] Further, to improve the temperature prediction accuracy of the model in the steel organization transformation stage, the embodiment of the present application can also introduce the phase change thermodynamic mechanism, and the latent heat generated by the plate in the cooling process is included in the energy balance calculation of the temperature evolution process, so as to realize the dynamic response and coupling of the phase change behavior under different plate types and different cooling rates.

[0224] As a specific example, the specific implementation steps of the above method are as follows:

[0225] (S151) latent heat modeling: for a specific steel type, the latent heat value in the phase change process from austenite to ferrite, bainite, pearlite or martensite is obtained through a thermodynamic database, literature or experimental method (differential scanning calorimetry DSC). Combined with the JMAK model and other dynamic equations, the phase change fraction changing with time is calculated, and the latent heat released or absorbed per unit time is obtained, which is used as a heat source term in the energy equation to participate in the temperature field solution.

[0226] (S152) integration of phase change kinetics model: combined with the continuous cooling transformation (CCT) curve or JMAK equation of the material, a phase change model that can be dynamically adjusted according to the actual cooling rate is established. For different types of phase change processes, diffusion-controlled phase change and shear-controlled phase change, Koistinen-Marburger equation and other models can be used for processing. The above model is coupled with the temperature field in an explicit manner to form an iterative solution mechanism, realizing real-time feedback of the phase change process and temperature evolution.

[0227] (S153) construction of steel parameterization database: a phase change database covering multiple common steel types (such as low carbon steel, medium carbon steel, high carbon steel, etc.) is established, including the relationship between density, specific heat capacity, thermal conductivity and temperature, the critical temperature interval and latent heat value of each phase change stage, and the phase change kinetics model parameters. The model can select the corresponding parameter set according to the actual production steel type for prediction.

[0228] (S154) coupling method with heat conduction model: the latent heat of phase change can be coupled with the energy equation through an explicit heat source term, simplifying the calculation process.

[0229] (S155) model verification and application expansion: the phase change temperature and organization evolution results of the model are verified and calibrated through experimental determination (such as thermal dilatometer, DSC test, metallographic analysis, etc.), to ensure the applicability of different steel types under different cooling paths. Further, the phase change model can be combined with the microstructure prediction and mechanical property regression model to build a multi-scale correlation path from temperature to organization to performance, providing more comprehensive decision support for hot rolling process control.

[0230] Through the above-mentioned embodiments, the capability blind area of the traditional temperature prediction control model in the phase change stage can be effectively supplemented, and the adaptability of the model to complex cooling behaviors is enhanced, which is particularly suitable for high-strength steel, alloy steel and other product process scenes with strict temperature control requirements, and further improves the product performance consistency and process control precision.

[0231] The traditional temperature control method usually relies on a fixed cooling strategy, lacks real-time prediction and feedback adjustment capability for temperature changes of the plate under complex process conditions, and causes poor finishing mill inlet temperature stability and large product performance fluctuations. Through the finishing mill inlet temperature prediction control method provided in the above embodiments, the technical problems of low finishing mill inlet temperature control precision and difficulty in adapting to temperature fluctuations caused by dynamic changes of different material and thickness plates in the hot rolling process are solved.

[0232] Further, through the method provided in the above embodiments, the present application achieves the following technical effects:

[0233] 1. Improve the finishing mill inlet temperature control precision: by constructing a temperature prediction control model based on historical data and dynamically adjusting the model parameters combined with real-time data, the temperature prediction is more close to the actual working condition, thereby improving the control precision.

[0234] 2. Enhance adaptability and robustness: for plates of different materials, thicknesses and cooling conditions, the control strategy can be automatically identified and adjusted to improve adaptability under varying process conditions.

[0235] 3. Optimize the running efficiency of the cooling system: based on the deviation between the model prediction result and the target temperature distribution, dynamically adjust the parameters of the cooling system to realize intelligent allocation and efficient use of cooling resources.

[0236] 4. Improve product quality stability: by accurately controlling the finishing mill inlet temperature, the uneven organization performance caused by temperature fluctuations is reduced, which helps to improve the mechanical properties and surface quality of the hot-rolled plate.

[0237] In summary, the present application provides a finishing mill inlet temperature prediction control method with self-learning ability and dynamic feedback adjustment mechanism, which significantly improves the automation and intelligent level in the hot rolling production process.

[0238] Based on the same inventive concept, the present application also provides a finishing mill inlet temperature prediction control device, as shown in Figure 3 The device comprises:

[0239] The model construction unit 310 is configured to construct a temperature prediction control model based on historical data in the hot rolling and cooling process of different plates, and the historical data includes the material quality, plate thickness and temperature change data under the corresponding cooling condition of the plate transmitted from the rough rolling outlet to the finishing mill inlet.

[0240] The model optimization unit 320 is configured to dynamically adjust parameters of the temperature prediction control model according to the real-time collected temperature data and the output result of the temperature prediction control model.

[0241] The temperature control unit 330 is configured to adjust parameters of the cooling system to control the cooling temperature at the finish rolling inlet based on a deviation between the temperature distribution predicted by the temperature prediction control model and a target temperature distribution.

[0242] The finish rolling inlet temperature prediction control device provided by the embodiments of the present application can be specific hardware on the device or software or firmware installed on the device. The device provided by the embodiments of the present application has the same implementation principle and technical effects as the foregoing method embodiments. For brevity and conciseness, the device embodiment part is not mentioned in the foregoing description. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the system, device and unit described above can be referred to the corresponding processes in the foregoing method embodiments, which will not be described herein. The finish rolling inlet temperature prediction control device provided by the embodiments of the present application has the same technical features as the finish rolling inlet temperature prediction control method provided by the foregoing embodiments, so it can also solve the same technical problems and achieve the same technical effects.

[0243] The embodiments of the present application also provide an electronic device, specifically, the electronic device includes a processor and a storage device; the storage device stores a computer program, and the computer program performs the method according to any one of the foregoing embodiments when executed by the processor.

[0244] Figure 4 The electronic device provided by the embodiments of the present application has a structure diagram as shown in FIG. 4. The electronic device 400 includes a processor 40, a memory 41, a bus 42 and a communication interface 43, and the processor 40, the communication interface 43 and the memory 41 are connected through the bus 42. The processor 40 is configured to execute executable modules stored in the memory 41, such as a computer program.

[0245] The memory 41 can include a high-speed random access memory (RAM) and can also include a non-volatile memory, such as at least one disk memory. The communication between the system network element and at least one other network element is realized through at least one communication interface 43 (which can be wired or wireless), and the Internet, a wide area network, a local area network, a metropolitan area network, etc. can be used.

[0246] The bus 42 can be an ISA bus, a PCI bus or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation,Figure 4 Only one bidirectional arrow is shown for each bus, but it will be understood that the bus is bidirectional, and that only the single bidirectional arrow is shown for clarity.

[0247] The memory 41 is configured to store a program, and the processor 40 executes the program after receiving an execution instruction. The method performed by the device for defining the flow according to any of the embodiments of the present application can be applied to the processor 40 or implemented by the processor 40.

[0248] The processor 40 can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor 40 or an instruction in the form of software. The processor 40 described above can be a general processor, including a central processing unit (CPU), a network processor (NP), and the like; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. Each method, step and logic block diagram disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, and the like storage medium mature in the art. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and combines the hardware to complete the steps of the above method.

[0249] Corresponding to the above method, the embodiment of the present application further provides a computer readable storage medium, the computer readable storage medium stores computer executable instructions, when the processor calls and runs the computer executable instructions, the computer executable instructions make the processor run the steps of the above method.

[0250] In the embodiments of the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. The embodiments described above are merely exemplary, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation; for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0251] In addition, the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0252] In addition, the functional modules in the various embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0253] It should be noted that if the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that make contributions to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0254] In this document, the terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations.

[0255] The above merely provides an example of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for predicting and controlling the inlet temperature of a finishing mill, characterized in that, The method includes: A temperature prediction and control model is constructed based on historical data from the hot rolling and cooling process of different plates. The historical data includes the material, thickness, and temperature change data of the plates transferred from the roughing mill exit to the finishing mill entrance under the corresponding cooling conditions. Based on the real-time collected temperature data and the output results of the temperature forecast control model, the parameters of the temperature forecast control model are dynamically adjusted. Based on the deviation between the temperature distribution predicted by the temperature prediction and control model and the target temperature distribution, the parameters of the cooling system are adjusted to control the cooling temperature at the finishing mill inlet. The input parameters of the temperature prediction and control model include: primary rolling exit temperature, temperature learning coefficient, time difference, radiation ratio, specific heat, and specific gravity; the output parameters of the temperature prediction and control model include the finishing rolling inlet temperature; the temperature prediction and control model predicts the temperature distribution of each plate during the current cooling process by calculating the average value of the finishing rolling inlet temperature. Based on real-time collected temperature data and the output of the temperature forecasting and control model, the parameters of the temperature forecasting and control model are dynamically adjusted, including: The predicted finishing mill inlet temperature output by the temperature prediction and control model is compared with the actual measured finishing mill inlet temperature, and the temperature deviation between the two is calculated. Based on the magnitude and trend of the temperature deviation, a feedback control algorithm is used to dynamically correct the temperature learning coefficient. When the temperature deviation exceeds a preset threshold, at least one of the input parameters, namely ambient temperature, radiation ratio, specific heat, or time difference, is recalibrated.

2. The finishing mill inlet temperature prediction and control method according to claim 1, characterized in that, The method for obtaining the input parameters of the temperature forecast control model includes: The initial rolling exit temperature of the plate was determined using data collected by an infrared pyrometer. The temperature learning coefficient is determined based on the ambient temperature, plate temperature, and stage coefficient at different cooling stages. The time difference is determined by the first timestamp when the temperature acquisition point of the plate reaches the primary rolling exit and the second timestamp when it reaches the finishing rolling inlet; Using a preset model relating the emissivity to the properties of the plate, the emissivity value of the plate at the current cooling stage is determined. Based on the specific heat versus temperature relationship curve determined in advance through experiments, the specific heat value of the plate at the current cooling stage is determined; The specific gravity of the board is determined based on its material.

3. The finishing mill inlet temperature prediction and control method according to claim 1, characterized in that, Based on the deviation between the temperature distribution predicted by the temperature prediction and control model and the target temperature distribution, the parameters of the cooling system are adjusted to control the cooling temperature at the finishing mill inlet, including: The adjusted temperature forecast control model is used to predict the temperature distribution during the current cooling process. The predicted temperature distribution is compared with the set target temperature distribution; Based on the comparison results exceeding the standard temperature deviation range, a parameter adjustment strategy for the cooling system is determined to control the cooling temperature at the finishing mill inlet.

4. The finishing mill inlet temperature prediction and control method according to claim 3, characterized in that, The parameters of the cooling system include: cooling air velocity, cooling time, and jet flow rate; the parameter adjustment strategy for the cooling system is determined, including: Based on the structure of the finishing mill inlet area, a three-dimensional geometric model of the cooling system is constructed; the cooling system includes nozzles disposed on the transport path of the plate, the nozzles being used to spray air-water atomization onto the plate; Based on the target temperature and preset cooling parameters of the current cooling stage, the cooling process of the plate is simulated using the three-dimensional geometric model. The parameter adjustment strategy of the cooling system is optimized based on the simulation results.

5. The finishing mill inlet temperature prediction and control method according to claim 4, characterized in that, The parameter adjustment strategy for the cooling system is optimized based on simulation results, including: Based on the simulation results, the effects of different nozzle layouts, injection speeds, and injection angles on the cooling uniformity and cooling efficiency of the plate were determined. Based on the impact analysis results, a cooling parameter adjustment strategy is generated, which includes: adjusting the distribution density and arrangement of nozzles on the transmission path of the plate; and adjusting the spray speed, spray angle and / or spray time of the nozzles.

6. A finishing mill inlet temperature prediction and control device, characterized in that, The apparatus is used to implement the control method according to any one of claims 1 to 5, the apparatus comprising: The model building unit is used to build a temperature prediction and control model based on historical data of different plates during hot rolling and cooling processes. The historical data includes the material, thickness, and temperature change data of the plates transferred from the roughing mill exit to the finishing mill entrance under the corresponding cooling conditions. The model optimization unit is used to dynamically adjust the parameters of the temperature forecast control model based on the real-time collected temperature data and the output results of the temperature forecast control model. The temperature control unit is used to adjust the parameters of the cooling system based on the deviation between the temperature distribution predicted by the temperature prediction control model and the target temperature distribution, so as to control the cooling temperature at the finishing mill inlet.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium stores computer-executable instructions, which, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 5.

Images