Multi-target real-time adaptive control method for cooling process after rolling

Abstract

The invention discloses a multi-target real-time self-adaptive control method for a cooling process after rolling. The method comprises the following steps: step 1, dividing a whole cooling area after rolling into a plurality of cooling sections and introducing a plurality of heat flux density self-adaptive coefficients; 2, a heat transfer equation of the strip steel cooling process is established, and the heat flow density self-adaption coefficient is introduced into the heat transfer equation; 3, after the strip steel enters the cooling area, new strip steel samples are created every fixed time step length, and the states of all the strip steel samples in the cooling area are updated; 4, after a certain strip steel sample passes through a certain pyrometer, real-time self-adaptive calculation is carried out according to the deviation between the actually measured temperature and the calculated temperature of the strip steel sample, and the heat flux density self-adaptive coefficient of the corresponding cooling section is updated; 5, after a certain strip steel sample passes through the coiling pyrometer, deleting the strip steel sample; and step 6, repeating the step 3 to the step 5 until the tail part of the strip steel passes through the coiling pyrometer, and ending the calculation.

Description

Technical Field

[0001] The invention belongs to the technical field of hot rolling and relates to a multi-objective real-time adaptive control method for a post-rolling cooling process. Background Art

[0002] As the last major production process of the hot rolling production line, post-rolling cooling is mainly used to cool the strip from the final rolling temperature of about 820-950℃ to the coiling temperature of 400-680℃ through a preset cooling path. In this way, hot-rolled products with expected microstructure and mechanical properties are obtained. However, the heat transfer between the strip and the surrounding medium in the cooling zone is very complicated during the cooling process, and it is difficult to quantitatively describe the process with high precision using only a simplified heat transfer model. Therefore, in actual production, in order to improve the control accuracy of post-rolling cooling temperature, model adaptive strategy is a common means to improve the calculation accuracy of the model. Advanced adaptive strategies can accurately and timely correct model calculation deviations, thereby improving the accuracy and robustness of the model. Therefore, adaptive strategies are crucial to the high-precision control of post-rolling cooling temperature and the quality of hot-rolled products.

[0003] In order to improve the temperature control accuracy of the post-rolling cooling process, domestic scholars have conducted in-depth research on the existing post-rolling cooling control system, and have successively developed adaptive strategies suitable for different control systems, achieving the improvement of the coiling temperature control accuracy of the post-rolling cooling process. The Chinese patent "CN 105921522B Laminar cooling temperature adaptive control method based on RBF neural network" designs a laminar cooling temperature adaptive control technology based on RBF neural network. By creating and evaluating the RBF neural network to obtain the latest adaptive parameters, the laminar cooling model will calculate the cooling strategy based on the latest adaptive parameters. The Chinese patent "CN102284517A Long-term self-learning method for post-rolling cooling based on case reasoning" is based on a large amount of on-site production data. Through case reasoning techniques such as case construction, case retrieval, case reuse, and case correction, the long-term self-learning coefficient in the mathematical model of control cooling is decided, so that the post-rolling cooling model has the ability to adapt to changes in working conditions, and significantly improves the head setting accuracy of the model. The Chinese patent "CN102284511A A temperature adaptive method for laminar cooling of steel strip" introduces different adaptive coefficients for the head, middle and tail of the steel strip, improves the temperature uniformity of the steel strip in the length direction, makes the self-learning system more perfect, and further improves the hit rate of coiling temperature. The Chinese patent "CN111814861B An online cooling control method based on dual self-learning models" combines the characteristics of unsupervised spatial clustering models and supervised deep neural network prediction models, complements each other's advantages, and establishes a system structure with parallel and weight-sharing dual self-learning models. Under low-cost data volume conditions, it realizes rapid combined self-learning, completes nonlinear coupling calculations in a short time, and improves the robustness and learning efficiency of the overall cooling control system. The Chinese patent "CN105032951B A control method for improving the accuracy and self-learning efficiency of ultra-fast cooling temperature models" comprehensively considers the influence of factors such as steel grade physical properties, steel plate specifications, medium water temperature, steel plate temperature, and aging on the heat transfer coefficient of the temperature model. A multi-dimensional spatial relationship is constructed to effectively describe the relationship between the heat transfer coefficients under the influence of different factors, quickly and accurately predict the value of the heat transfer coefficient under the target influence conditions, and finally achieve precise control of the temperature model. The Chinese patent "CN102794315B A self-learning method for improving the prediction accuracy of the full-length coiling temperature of strip steel" aims at the hysteresis problem existing in the inter-segment self-learning method in the coiling temperature control of strip steel. It uses the inter-segment self-learning related control parameters of the rolled strip steel to calculate the hysteresis factor of the inter-segment self-learning, compensate for the hysteresis effect of the current inter-segment self-learning of the strip steel, and thus improve the coiling temperature prediction accuracy of each section of the full length of the strip steel.

[0004] The above research has promoted the accuracy of post-rolling cooling temperature control to a certain extent, but it is more of an adaptive method designed to improve the accuracy of coiling temperature control. The entire cooling zone shares the same adaptive coefficient and the adaptive operation is always triggered after the strip head runs to the coiling pyrometer, resulting in poor control accuracy of more than 100 meters of the strip head due to the inability to adapt. And when only one adaptive coefficient is used, overshoot and multi-objective adjustment conflicts are prone to occur. The so-called multi-objective adjustment conflict problem refers to when there are multiple control targets, the adjustment directions of each control target are opposite. For example, in the post-rolling cooling process, the temperature at the middle position of the cooling zone and the coiling temperature need to be controlled at the same time. At a certain moment, the measured temperature at the middle position is greater than the model calculated temperature, and the adaptive coefficient needs to be reduced. The measured value of the coiling temperature is less than the model calculated value. In order to ensure the accuracy of the coiling temperature calculation, the adaptive coefficient needs to be increased. At this time, if there is only one adaptive coefficient, there will be an adjustment conflict problem. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a multi-objective real-time adaptive control method for the post-rolling cooling process, by dividing the entire cooling zone into multiple cooling sections, introducing an adaptive coefficient corresponding to each cooling section, and correcting the adaptive coefficient of the section in real time according to the measured temperature at the outlet of each section and the temperature calculated by the model. This improves the calculation accuracy of the model for the strip head and solves the problem of overshoot and multi-objective adjustment conflict in the traditional adaptive coefficient.

[0006] The present invention provides a multi-objective real-time adaptive control method for a post-rolling cooling process, comprising: Step 1: Divide the entire post-rolling cooling zone into multiple cooling sections and introduce multiple heat flux adaptive coefficients; Step 2: Establish the heat transfer equation of the strip cooling process and introduce the heat flux adaptive coefficient into the heat transfer equation; Step 3: When the strip enters the cooling zone, a new strip sample is created every fixed time step, and the status of all strip samples in the cooling zone is updated; Step 4: After a steel strip sample passes through a pyrometer, a real-time adaptive calculation is performed based on the measured temperature and calculated temperature deviation of the steel strip sample to update the heat flux adaptive coefficient of the corresponding cooling section; Step 5: After a strip steel sample passes through the coiling pyrometer, the strip steel sample is deleted; Step 6: Repeat steps 3 to 5 until the tail of the strip passes through the coiling pyrometer and the calculation is completed.

[0007] The present invention provides a multi-objective real-time adaptive control method for a post-rolling cooling process. The method divides the entire post-rolling cooling area into a plurality of cooling sections and introduces a plurality of adaptive coefficients. Each adaptive coefficient is adjusted in real time by the deviation between the measured temperature at the outlet of the corresponding cooling section and the calculated temperature. This enables the model calculation deviation to be corrected in a very short time, greatly reduces the adaptive delay time, solves the multi-objective adjustment conflict problem existing in the traditional adaptive method, and thus improves the temperature control accuracy of the post-rolling cooling process. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a flow chart of a multi-objective real-time adaptive control method for a post-rolling cooling process of the present invention; Figure 2 This is a distribution diagram of post-rolling cooling equipment in an embodiment of the present invention; Figure 3 is the actual measured temperature of the strip steel sample after final rolling in the embodiment of the present invention; Figure 4 This is a comparison diagram of ultrafast cooling temperature using the traditional adaptive method and the multi-objective real-time adaptive method of the present invention; Figure 5 The comparison diagram of the intermediate temperature using the traditional adaptive method and the multi-objective real-time adaptive method of the present invention; Figure 6 The figure is a comparison chart of coiling temperatures using the traditional adaptive method and the multi-objective real-time adaptive method of the present invention. DETAILED DESCRIPTION

[0009] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0010] like Figure 1 As shown, a multi-objective real-time adaptive control method for a post-rolling cooling process of the present invention comprises: Step 1: Divide the entire post-rolling cooling zone into multiple cooling sections and introduce multiple heat flux adaptive coefficients, specifically: Step 1.1: Based on the number of pyrometers in the post-rolling cooling zone n The entire post-rolling cooling zone is divided into n- 1 cooling section, the starting position and the ending position of each cooling section are the positions of the pyrometers.

[0011] Step 1.2: According to the number of cooling sections divided in step 1.1, introduce n- 1 initial heat flux density adaptive coefficient, the initial heat flux density adaptive coefficient is set according to the empirical value.

[0012] The post-rolling cooling equipment of this production line is distributed as follows: Figure 2 As shown in the figure, four pyrometers are installed at different positions in the post-rolling cooling area of ​​the production line, which are used to detect the final rolling temperature FDT, the ultra-fast cooling outlet temperature UFCT, the middle position temperature MT of the cooling area and the temperature before the coiler CT, which are represented by 1#PY, 2#PY, 3#PY and 4#PY. Therefore, the post-rolling cooling area is divided into three cooling sections, and three adaptive coefficients are introduced. The effective area of ​​each adaptive coefficient is shown in Figure 2 shown.

[0013] Step 2: Establish the heat transfer equation of the strip cooling process and introduce the heat flux adaptive coefficient into the heat transfer equation, specifically: Step 2.1: Since the width-to-thickness ratio of the strip steel is relatively large during the cooling process, the three-dimensional heat transfer is simplified to a one-dimensional heat transfer problem along the thickness direction, and a one-dimensional unsteady-state heat conduction differential equation is established, as shown in the following equation: (1) Starting conditions: (2) Boundary conditions: (3) In the formula, ρ is the density of the strip, in kg / m 3 ; c p is the specific heat capacity, in J / (kg·℃); λ is thermal conductivity, unit is W / (m·℃); T is the strip temperature, in °C; x is the thickness direction coordinate, t is the cooling time, in units of s ; h is the strip thickness, in m; T (0, h ) is the final rolling temperature, in °C; k i For the i The heat flux adaptive coefficient of each cooling section is: ; α t and α b are the convective heat transfer coefficients of the upper and lower surfaces, respectively, in W / (m²·℃); T h and T 0 are the upper and lower surface temperatures of the strip, in °C; T e It is the temperature of cooling medium around the strip, in °C.

[0014] Step 2.2: Divide the strip into N The Crank-Nicholson method is used to solve equation (1) by difference, and the temperature distribution in the thickness direction of the strip at any time is obtained, as shown in the following equation: (4) In the formula, , , , , , , , , , , .in, a t , a b , b , c , d t , d b , e and f It is an intermediate variable defined for writing convenience and has no practical significance; B it and B ib are the Biot numbers of the upper and lower surfaces respectively, F 0 is the Fourier number.

[0015] Δ x is the distance between two grid nodes, in m; Δ t To calculate the time step, the value is 192ms; is the upper surface temperature of the strip sample at the current moment, is the lower surface temperature of the strip sample at the current moment, is the number of the strip sample at the current moment i The grid node temperature; the unit is ℃; is the upper surface temperature of the strip sample at the next moment, is the lower surface temperature of the strip sample at the next moment, is the number of the strip sample at the next moment i The grid node temperature, in °C; T e It is the temperature of cooling medium around the strip, in °C.

[0016] For the convenience of subsequent writing, equation (4) is simplified to the form of matrix multiplication, as shown below: (5) (6) (7) (8) in, is the temperature vector of the steel strip sample at the next moment, is the temperature vector of the steel strip sample at the current moment.

[0017] The thickness and thermophysical parameters of the steel strip in this embodiment are shown in Table 1. Since the convective heat transfer coefficient actually exists but cannot be measured, it is assumed that the convective heat transfer coefficient of each area in the cooling process and the time taken for the steel strip sample to pass through each area are shown in Table 2, where Actual represents the actual convective heat transfer coefficient, the temperature calculated using this coefficient as the boundary condition is the actual temperature, and Model is the convective heat transfer coefficient of each area calculated by the field heat transfer model.

[0018] Table 1 Parameters required for heat transfer process

[0019] Table 2 Equivalent convection heat transfer coefficients in different regions

[0020] Step 3: When the strip enters the cooling zone, a new strip sample is created every fixed time step (192ms), and the status of all strip samples in the cooling zone is updated.

[0021] The state of each strip sample includes its position in the cooling zone, cooling time, thickness, moving speed, first calculation temperature, second calculation temperature and third calculation temperature, which are respectively based on the current heat flux adaptive coefficient. Corrected convection heat transfer coefficient, using initial heat flux adaptive coefficient Corrected convective heat transfer coefficient and the use of The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector of the steel strip sample through equation (5). The step 3 is specifically as follows: Step 3.1: Collect the actual temperature measured by the final rolling pyrometer at the current moment as the starting condition for the strip sample to participate in subsequent calculations, and record the position of the strip sample at this moment as position 0, the cooling time is 0s, the moving speed and thickness are the current measured speed and thickness, and the initial values ​​of the three calculated temperatures are all the measured temperatures at this moment.

[0022] In this embodiment, the actual final rolling temperature of each strip steel sample is as follows: Figure 3 shown.

[0023] Step 3.2: Move all the strip steel samples in the cooling zone forward by one time step, and determine which heat flux adaptive coefficient should be used based on the location of the strip steel samples and whether they pass through the water cooling area during the movement.

[0024] Step 3.3: Update the status of all strip samples in the cooling zone according to the following formula: (9) (10) (11) (12) (13) In the formula, and Respectively j The current and next position of the strip steel sample, in m. , m is the number of strip samples; v is the current rolling speed, in m / s; and Respectively j The cooling time of the strip steel sample at the current moment and the next moment, in seconds; For the j The current heat flux adaptive coefficient is used for each steel strip sample at the current moment The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector through equation (5), For the j The initial heat flux adaptive coefficient is used for the current moment of the steel strip sample The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector through equation (5), For the j The strip steel samples are currently using The corrected convective heat transfer coefficient is used as the temperature vector calculated by equation (5) as the boundary condition; and For the j The current heat flux adaptive coefficient is used for the next moment of the steel strip sample The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector through equation (5), For the j The initial heat flux adaptive coefficient is used for the next moment of the steel strip sample The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector through equation (5), For the j The next moment, the strip steel sample is used The corrected convective heat transfer coefficient is used as the boundary condition for the temperature vector calculated by equation (5).

[0025] That is, in formula (11) , in formula (12) , in formula (13) ; Δ k is a fixed constant with a smaller value.

[0026] In this embodiment The value is 1.0, Δ k When the value is 0.05, the temperature calculated using the convective heat transfer coefficient in Table 2 as the boundary condition when the first strip steel sample passes through each high temperature meter is shown in Table 3.

[0027] Table 3 Temperatures at key locations during cooling process

[0028] Step 4: When a steel strip sample passes through a pyrometer, a real-time adaptive calculation is performed based on the measured temperature and calculated temperature deviation of the steel strip sample to update the heat flux adaptive coefficient of the corresponding cooling section, specifically: Step 4.1: Calculate the learning rate at the current moment according to the position of the current strip sample in the cooling zone and its length from the strip head, as follows: (14) In the formula, is the learning rate at the current moment; is the initial learning rate, which is a fixed constant; For the j The current position of the strip steel sample in the cooling zone, l j For the j The length of a strip sample from the strip head at the current moment, in meters.

[0029] In this embodiment, The value is 0.1; the positions of the four pyrometers in the post-rolling cooling zone are 0m, 23.56m, 56.24m and 110.8m respectively. Therefore, when the first sample passes through 2#PY, the position of the sample in the cooling zone is the position of 2#PY, that is, 23.56m. Because it is the first sample, its distance from the strip head is 0m. Therefore, the self-learning coefficient at this time should be:

[0030] Step 4.2: According to the deviation between the measured temperature and the calculated temperature of the strip sample, the heat flux adaptive coefficient of the corresponding cooling section is updated by the following formula: (15) In the formula, Updated for the next moment i The value of the heat flux adaptive coefficient of each cooling section; For the current moment i The value of the heat flux adaptive coefficient of each cooling section; Δ k is a fixed constant; T m is the measured temperature of the strip steel sample, in °C; For the j The current adaptive coefficient is used for each steel strip sample at the current moment The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector in equation (5). N The node temperature, For the j The initial adaptive coefficient is used for the current moment of the steel strip sample The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector in equation (5). N The node temperature, For the j The strip steel samples are currently using The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector in equation (5). N Node temperature.

[0031] Step 5: After a strip steel sample passes through the coiling pyrometer, the strip steel sample is deleted; , , , The values ​​of are 792.6, 779.3, 779.3 and 774.4℃ respectively, then the adaptive coefficient of the first cooling section at the next moment is:

[0032] Step 6: Repeat steps 3 to 5 until the tail of the strip passes through the coiling pyrometer and the calculation is completed.

[0033] The calculated temperature of each target in the real-time adaptive method using a single adaptive coefficient in this embodiment and the multi-target real-time adaptive method using multiple adaptive coefficients in the present invention is as follows: Figure 4~Figure 6 As shown in the figure, it can be clearly seen that the multi-objective real-time adaptive method proposed by the present invention can effectively solve the overshoot problem existing in the traditional adaptive method. The calculated temperature at each position is closer to the actual temperature, the calculation deviation is smaller, and the calculation accuracy of the model can be effectively improved. And because the adjustment target of each position is adjusted by a separate adaptive coefficient, the multi-objective adjustment conflict problem existing in the traditional one adaptive coefficient is solved.

[0034] The above description is only a preferred embodiment of the present invention and is not intended to limit the concept of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-objective real-time adaptive control method for post-rolling cooling process, characterized in that: include: Step 1: Divide the entire post-rolling cooling zone into multiple cooling sections and introduce multiple heat flux adaptive coefficients; Step 2: Establish the heat transfer equation of the strip cooling process and introduce the heat flux adaptive coefficient into the heat transfer equation; Step 3: When the strip enters the cooling zone, a new strip sample is created every fixed time step, and the status of all strip samples in the cooling zone is updated; Step 4: After a steel strip sample passes through a pyrometer, a real-time adaptive calculation is performed based on the measured temperature and calculated temperature deviation of the steel strip sample to update the heat flux adaptive coefficient of the corresponding cooling section; Step 5: After a strip steel sample passes through the coiling pyrometer, the strip steel sample is deleted; Step 6: Repeat steps 3 to 5 until the tail of the strip passes through the coiling pyrometer and the calculation is completed.

2. A multi-objective real-time adaptive control method for a post-rolling cooling process according to claim 1, characterized in that: The step 1 is specifically as follows: Step 1.1: Based on the number of pyrometers in the post-rolling cooling zone n The entire post-rolling cooling zone is divided into n- 1 cooling section, the starting and ending positions of each cooling section are the positions of the pyrometers; Step 1.2: According to the number of cooling sections divided in step 1.1, introduce n- 1 initial heat flux density adaptive coefficient, the initial heat flux density adaptive coefficient is set according to the empirical value.

3. A multi-objective real-time adaptive control method for a post-rolling cooling process as claimed in claim 2, characterized in that: The step 2 is specifically as follows: Step 2.1: Since the width-to-thickness ratio of the strip steel is relatively large during the cooling process, the three-dimensional heat transfer is simplified to a one-dimensional heat transfer problem along the thickness direction, and a one-dimensional unsteady-state heat conduction differential equation is established, as shown in the following equation: Starting conditions: Boundary conditions: In the formula, ρ is the density of the strip, in kg / m 3 ; c p is the specific heat capacity, in J / (kg·℃); λ is thermal conductivity, unit is W / (m·℃); T is the strip temperature, in °C; x is the thickness direction coordinate, t is the cooling time, in units of s ; h is the strip thickness, in m; T (0, h ) is the final rolling temperature, in °C; k i For the i The heat flux adaptive coefficient of each cooling section is: ; α t and α b are the convective heat transfer coefficients of the upper and lower surfaces, respectively, in W / (m²·℃); T h and T 0 are the upper and lower surface temperatures of the strip, in °C; T e is the temperature of the cooling medium around the strip, in °C; Step 2.2: Divide the strip into N The one-dimensional unsteady-state heat conduction differential equation in step 2.1 is solved by difference using the Crank-Nicholson method to obtain the temperature distribution in the thickness direction of the strip at any time, as shown in the following formula: In the formula, , , , , , , , , , , ;in, a t , a b , b , c , d t , d b , e and f It is an intermediate variable defined for writing convenience and has no practical significance; B it and B ib are the Biot numbers of the upper and lower surfaces respectively, F 0 is the Fourier number; Δ x is the distance between two grid nodes, in m; Δ t To calculate the time step, the value is 192ms; is the upper surface temperature of the strip sample at the current moment, is the lower surface temperature of the strip sample at the current moment, is the number of the strip sample at the current moment i The grid node temperature; the unit is ℃; is the upper surface temperature of the strip sample at the next moment, is the lower surface temperature of the strip sample at the next moment, is the number of the strip sample at the next moment i The grid node temperature, in °C; T e is the temperature of the cooling medium around the strip, in °C; For the convenience of subsequent writing, the temperature distribution formula in the thickness direction of the strip is simplified into the form of matrix multiplication, as shown below: in, is the temperature vector of the steel strip sample at the next moment, is the temperature vector of the steel strip sample at the current moment.

4. A multi-objective real-time adaptive control method for a post-rolling cooling process according to claim 1, characterized in that: The state of each strip sample includes its position in the cooling zone, cooling time, thickness, moving speed, first calculation temperature, second calculation temperature and third calculation temperature, which are respectively based on the current heat flux adaptive coefficient. Corrected convection heat transfer coefficient, using initial heat flux adaptive coefficient Corrected convective heat transfer coefficient and the use of The temperature vector of the steel strip sample is calculated by using the simplified temperature distribution formula in the thickness direction of the steel strip using the modified convective heat transfer coefficient as the boundary condition. The step 3 is specifically as follows: Step 3.1: The actual temperature measured by the final rolling pyrometer at the current moment is collected as the starting condition for the strip sample to participate in the subsequent calculation, and the position of the strip sample at this moment is recorded as position 0, the cooling time is 0s, the moving speed and thickness are the current measured speed and thickness, and the initial values ​​of the three calculated temperatures are all the measured temperatures at this moment; Step 3.2: Move all the strip steel samples in the cooling zone forward by one time step, and determine which heat flux adaptive coefficient should be used according to the location of the strip steel samples and whether they pass through the water cooling area during the movement; Step 3.3: Update the status of all strip samples in the cooling zone according to the following formula: In the formula, and Respectively j The current and next position of the strip steel sample, in m. , m is the number of strip samples; v is the current rolling speed, in m / s; and Respectively j The cooling time of the strip steel sample at the current moment and the next moment, in seconds; For the j The current heat flux adaptive coefficient is used for each steel strip sample at the current moment The temperature vector is calculated by using the simplified temperature distribution formula in the thickness direction of the strip using the modified convection heat transfer coefficient as the boundary condition. For the j The initial heat flux adaptive coefficient is used for the current moment of the steel strip sample The temperature vector is calculated by using the simplified temperature distribution formula in the thickness direction of the strip using the modified convection heat transfer coefficient as the boundary condition. For the j The strip steel samples are currently using The temperature vector calculated by the simplified temperature distribution formula in the thickness direction of the strip using the modified convective heat transfer coefficient as the boundary condition; and For the j The current heat flux adaptive coefficient is used for the next moment of the steel strip sample The temperature vector is calculated by using the simplified temperature distribution formula in the thickness direction of the strip using the modified convection heat transfer coefficient as the boundary condition. For the j The initial heat flux adaptive coefficient is used for the next moment of the steel strip sample The temperature vector is calculated by using the simplified temperature distribution formula in the thickness direction of the strip using the modified convection heat transfer coefficient as the boundary condition. For the j The next moment, the strip steel sample is used The temperature vector, Δ k is a fixed constant.

5. A multi-objective real-time adaptive control method for a post-rolling cooling process according to claim 1, characterized in that: The step 4 is specifically as follows: Step 4.1: Calculate the learning rate at the current moment according to the position of the current strip sample in the cooling zone and its length from the strip head, as follows: In the formula, is the learning rate at the current moment; is the initial learning rate, which is a fixed constant; For the j The current position of the strip steel sample in the cooling zone, l j For the j The current distance of each strip sample from the strip head, in meters; Step 4.2: According to the deviation between the measured temperature and the calculated temperature of the strip sample, the heat flux adaptive coefficient of the corresponding cooling section is updated by the following formula: In the formula, Updated for the next moment i The value of the heat flux adaptive coefficient of each cooling section; For the current moment i The value of the heat flux adaptive coefficient of each cooling section; Δ k is a fixed constant; T m is the measured temperature of the strip steel sample, in °C; For the j The current adaptive coefficient is used for each steel strip sample at the current moment The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector by simplifying the temperature distribution formula in the thickness direction of the strip. N The node temperature, For the j The initial adaptive coefficient is used for the current moment of the steel strip sample The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector by simplifying the temperature distribution formula in the thickness direction of the strip. N The node temperature, For the j The strip steel samples are currently using The corrected convective heat transfer coefficient is used as the boundary condition to calculate the temperature vector by simplifying the temperature distribution formula in the thickness direction of the strip. N Node temperature.

Images