A temperature field real-time compensation and intelligent control method for a roller heat treatment process

Abstract

This invention discloses a real-time temperature field compensation and intelligent control method for the heat treatment process of rolls, belonging to the field of automatic control technology. The method includes: constructing a thermal coupling relationship matrix between different sections based on Fourier's law of thermal conductivity; constructing a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculating the convective heat transfer coefficient in real time and dynamically correcting the heating power; establishing a dynamic estimation model of the roll core temperature based on the lumped parameter method, solving the core temperature in real time and compensating for thermal inertia lag in advance; and solving the above models collaboratively, using the weighted sum of squares of surface temperature deviation and core temperature deviation as the objective function, to solve for the optimal heating power of each section under thermal coupling constraints. This invention, through the combined application of classical thermophysics theories, solves the technical problems of thermal inertia lag, large axial temperature difference, and difficulty in controlling the core-surface temperature difference during the heat treatment process of rolls, achieving precise layered and zoned control of the temperature field, and significantly improving the quality of heat treatment.

Description

Technical Field

[0001] This invention belongs to the field of automatic control technology, and in particular relates to a method for real-time compensation and intelligent control of temperature field in the heat treatment process of rolling mill rolls. Background Technology

[0002] Rolls are crucial components of rolling mills, and the quality of their heat treatment directly affects their service life and rolling accuracy. During the heat treatment of rolls, the uniformity and control precision of the temperature field are key process parameters. However, existing technologies face the following technical challenges in controlling the heat treatment temperature of rolls: First, as a component with a large heat capacity, the rolling mill roll has a significant thermal inertia effect during the heat treatment process, resulting in a lag in temperature response. Traditional PID control is difficult to balance response speed and overshoot suppression. Second, there is a temperature gradient between the roll surface and the core. Existing control methods usually only monitor the surface temperature, which cannot accurately reflect the overall thermal state of the roll, resulting in uneven core structure transformation. Third, the temperature field distribution inside the heat treatment furnace is uneven, and the heating conditions at different axial positions of the rolls are different. Existing technologies lack targeted zoning compensation mechanisms. Fourth, the temperature detection signal is affected by environmental interference and the dynamic response characteristics of the sensor. Direct feedback control is prone to oscillation, which reduces system stability. Summary of the Invention

[0003] This invention aims to provide a real-time temperature field compensation and intelligent control method for the heat treatment process of rolling mill rolls, to solve the technical problems in existing rolling mill roll heat treatment temperature control, such as thermal inertia lag, large axial temperature difference, and difficulty in controlling the core-surface temperature difference. The specific solution is as follows: A method for real-time temperature field compensation and intelligent control in a roll heat treatment process, the method specifically includes the following steps: S1. Establish an axial partitioned heat conduction model for the heat treatment process of the roll. Divide the roll into N equidistant control sections along the axial direction. Each section is equipped with an independent thermocouple temperature measuring point and an independent heating actuator. Construct the thermal coupling relationship matrix between each section based on Fourier's law of thermal conductivity. S2. Construct a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculate the convective heat transfer coefficient between the roll surface and the furnace environment in real time, and dynamically correct the output value of heating power in each section according to the deviation between the measured temperature of the roll surface and the set temperature. S3. Establish a dynamic estimation model for the core temperature of the roll based on the lumped parameter method. Utilize the measured temperature of the roll surface and the time constant to calculate the dynamic change trend of the core temperature of the roll in real time. When there is a deviation between the estimated core temperature and the target temperature, adjust the heating power in advance to compensate for thermal inertia hysteresis. S4. The above thermal coupling relationship matrix, boundary heat transfer dynamic compensation model and core temperature estimation model are solved in a coordinated manner to generate real-time power control commands for each section, and drive the heating actuator to perform layered and zoned coordinated heating control of the rolls.

[0004] Further, step S1 specifically includes: S1-1. Establish the heat balance equations for each section: Let the thermal conductivity of the roll material be λ, the density be ρ, the specific heat capacity be c, and the temperature at the center point of the i-th section be T. i Let Δx be the distance between the center of segment i and the center of segment i+1. Then the heat flux density q from segment i to segment i+1 is... i→i+1 For: q i→i+1 =-λ·(T i+1 -T i The heat balance equations for each section are: ) / Δx V i Let Q be the volume of the i-th segment. i The heating power A provided to the heating actuator of the i-th section i-1 Let A be the contact area between the (i-1)th segment and the ith segment. i Let be the contact area between the i-th segment and the (i+1)-th segment; S1-2. Discretize the above heat balance equation and write it in matrix form: Where C is the heat capacity diagonal matrix, K is the thermal conductivity coupling matrix, T is the temperature vector, and Q is the heating power vector; the off-diagonal elements k of the thermal conductivity coupling matrix K are... ij The coefficient representing the influence of segment j on the heat conduction of segment i is calculated using the formula k. ij =λA ij / △x ij A ij The contact area between adjacent sections, △x ij This represents the center-to-center distance between adjacent segments.

[0005] Furthermore, step S2 specifically includes: S2-1. Calculate the convective heat transfer q between the surface of the roll in the i-th section and the furnace environment. conv，i q conv，i =h i (T) s，i -T ∞，i ), where h i Let T be the convective heat transfer coefficient of the i-th segment. s，i Let T be the surface temperature of the roll in the i-th section. ∞，i Let be the furnace ambient temperature of the i-th section; S2-2, Based on the real-time calculated h i The value is used to dynamically correct the heating power compensation amount △Q in the i-th segment.i △Q i =h i A s，i (T) s，i -T set，i ), where A s，i Let T be the surface area of ​​the i-th segment. set，i Set the temperature for the surface of the roll in the i-th section.

[0006] Furthermore, the convective heat transfer coefficient h is calculated using the empirical formula: h = N u ·λ air / L; where N u λ is the Nusselt number, a dimensionless number representing the ratio of convective heat transfer intensity to conductive heat transfer intensity; air is the thermal conductivity of air, in W / (m·K), which can be obtained from a table; L is the characteristic dimension, which is usually the diameter of the roll for horizontally placed rolls, in meters. The Nusselt number N u The calculation uses the empirical correlation for natural convection heat transfer: N u =C·(G r ·P r ) n ; where G r The Grashof number is a dimensionless number representing the ratio of buoyancy to viscous force in natural convection; P r The Prandtl number is a dimensionless number that represents the ratio of momentum diffusion capacity to heat diffusion capacity; C and n are empirical constants. For natural convection in a horizontal cylinder, C is usually taken as 0.53 and n as 0.25 in laminar flow, and C is usually taken as 0.13 and n as 0.33 in turbulent flow. The formula for calculating the Grashof number Gr is: Gr = gβ(T) s -T ∞ L 3 / ν; where g is the acceleration due to gravity, taken as 9.8 m / s²; β is the coefficient of thermal expansion of air, for an ideal gas, β≈1 / T ∞ The unit is K. -1 ν is the kinematic viscosity of air, in m² / s, which can be obtained from a table; L is the characteristic dimension (same as above), in meters. The formula for calculating Prandtl's number Pr is: P r =ν / α; where α is the thermal diffusivity of air, in m² / s, α=λ air / (ρ air -c ρ，air ), where λ air For air density, c ρ，air This refers to the specific heat capacity of air at constant pressure.

[0007] Furthermore, step S3 specifically includes: S3-1. Let the core temperature of the roll in the i-th section be T. core，i (t) and its surface temperature T s，i (t) satisfies the following relation: (T) s，i (t)-T ∞，i ) / (T core，i (t)-T ∞，i ) = 1 - exp(-t / τ); where τ is the thermal time constant of the roll, and its calculation formula is: τ = ρcV i / h i A s，i V i Let A be the volume of the i-th section of the roll. s，i Let be the surface area of ​​the i-th section of the roll; S3-2, Based on the measured surface temperature sequence T of the i-th segment s，i (t) k The core temperature T of the i-th segment at the current time is estimated using a reverse solution method. core，i (t) k ), T core，i (t) k )=T ∞，i +(T) s，i (t) k )-T ∞，i ) / (1-exp(-t k / τ)), where t k This represents the k-th sampling time. S3-3, When the estimated core temperature T of the i-th segment... core，i (t) k ) and the target core temperature T of the i-th segment core，i，target When a deviation exists, calculate the advance compensation power ΔQ of the i-th segment. core，i ΔQ core，i =ρcV i / △t·(T core，i，target -T core，i (t) k )); where Δt is the control period.

[0008] Further, step S4 includes constructing the comprehensive control objective function J: ; Among them, w s,i w represents the weighting coefficient for the surface temperature deviation of the i-th segment, reflecting the importance attached to the surface temperature control accuracy of this segment; core T is the weighting coefficient for core temperature deviation, reflecting the importance attached to the accuracy of core temperature control; core，setThe average core temperature estimate for the entire roll (T for each section) core，i (t) k (arithmetic mean of T) core，target The target core temperature is set uniformly for the entire roll.

[0009] Furthermore, the value of N ranges from 3 to 20. The thermocouple temperature measuring point configured independently for each section is located on the surface of the roll at the center of the section. The heating actuator configured independently for each section is an induction heating coil or a resistance heating element. The furnace ambient temperature T in section i ∞，i The method of obtaining the temperature is as follows: an independent ambient temperature sensor is set in the furnace space corresponding to each control section i to collect the local ambient temperature of the section in real time, which is used for the accurate calculation of the boundary heat transfer compensation model of the section.

[0010] Furthermore, the thermal time constant τ of the roll is dynamically updated using a segmented identification method: the heat treatment process is divided into a heating stage, a holding stage, and a cooling stage, and the thermal time constant τ of each stage is identified separately. heat τ hold and τ cool Used for core temperature estimation at each stage; The weighting coefficient w s，i and w core，i A dynamic adjustment strategy is adopted: during the heating stage, the weight of the core temperature deviation w is increased. core，i To accelerate core heating; during the heat preservation stage, increase the weighting of surface temperature deviation w. s，i To ensure uniform surface temperature; during the cooling phase, the weights of the two are balanced to achieve uniform cooling.

[0011] Furthermore, it also includes S5, an adaptive protection step for abnormal operating conditions: when the measured surface temperature T of any section... s，i With the set temperature T set，i The absolute value of the deviation exceeds the preset threshold ΔT max When this occurs, the abnormal protection mechanism is triggered, reducing the heating power Q of this section. i Force the value to zero and issue an alarm signal.

[0012] Another embodiment of the present invention provides a real-time temperature field compensation and intelligent control system for the heat treatment process of rolls, including a first module, a second module, a third module and a fourth module, which are connected in sequence. The first module is used to establish an axial partitioned heat conduction model for the heat treatment process of the roll, dividing the roll into N equidistant control sections along the axial direction. Each section is equipped with an independent thermocouple temperature measuring point and an independent heating actuator. Based on Fourier's law of thermal conductivity, a thermal coupling relationship matrix between each section is constructed. The second module is used to construct a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculate the convective heat transfer coefficient between the roll surface and the furnace environment in real time, and dynamically correct the output value of heating power in each section according to the deviation between the measured temperature of the roll surface and the set temperature. The third module is used to establish a dynamic estimation model of roll core temperature based on the lumped parameter method. It uses the measured temperature of the roll surface and the time constant to calculate the dynamic change trend of the roll core temperature in real time. When there is a deviation between the estimated core temperature and the target temperature, the heating power is adjusted in advance to compensate for thermal inertia hysteresis. The fourth module is used to collaboratively solve the above-mentioned thermal coupling relationship matrix, boundary heat transfer dynamic compensation model and core temperature estimation model, generate real-time power control commands for each section, and drive the heating actuator to perform layered and zoned collaborative heating control of the rolls.

[0013] Compared with the prior art, the present invention has at least one of the following technical effects: 1. This invention establishes a dynamic core temperature estimation model based on the lumped parameter method, using the thermal time constant to describe the thermal inertia of the roll, and calculating the core temperature change trend in real time based on the surface temperature. When there is a deviation between the estimated core temperature and the target, the advance compensation power is calculated in advance, and the heating is adjusted before the surface temperature deviates significantly (in traditional PID control, the controller only adjusts the feedback based on the surface temperature deviation. Due to the large heat capacity and long heat conduction path of the roll, it takes a long time for the surface temperature change to be transmitted to the core. When the surface temperature deviates from the set value, the core temperature may have already deviated more significantly. At this time, it is too late to adjust the heating power, resulting in temperature overshoot or undershoot, which seriously affects the heat treatment quality). This invention changes passive feedback to active feedforward, effectively suppressing temperature overshoot and ensuring that the core microstructure transformation is sufficient and uniform.

[0014] 2. This invention divides the roll into multiple independent control sections and constructs a thermal coupling matrix based on Fourier's law of thermal conductivity to accurately describe the heat transfer between adjacent sections. In the optimization solution, the thermal coupling matrix is ​​used as a constraint condition to ensure that the heating power of each section not only meets the control requirements of its own section but also takes into account the influence of adjacent sections, thereby significantly reducing the axial temperature difference and significantly improving the uniformity of the microstructure properties along the length of the roll.

[0015] 3. This invention simultaneously introduces surface temperature deviation and core temperature deviation terms into the objective function, achieving layered collaborative control through dynamic adjustment of weighting coefficients. During the heating phase, the core weight is increased to accelerate heat penetration; during the heat preservation phase, the surface weight is increased to ensure uniformity; and during the cooling phase, both are balanced to control thermal stress. This effectively reduces the core-surface temperature difference and improves the uniformity of roll cross-sectional hardness, significantly extending service life.

[0016] 4. This invention is based on Newton's law of cooling and the theory of natural convection heat transfer. In each control cycle, the convection heat transfer coefficient is calculated in real time based on the measured temperature, and the heating power compensation is dynamically adjusted. Independent ambient temperature sensors are installed in each section to eliminate the influence of uneven furnace temperature on heat transfer calculations. This significantly improves the system's adaptability to changes in the furnace environment and greatly enhances control accuracy.

[0017] 5. This invention divides the heat treatment process into three stages: heating, holding, and cooling. It identifies the thermal time constant for each stage and dynamically updates the core temperature estimation model. This solves the model mismatch problem caused by changes in material thermal properties with temperature, significantly reducing core temperature estimation errors and providing an accurate basis for proactive compensation control.

[0018] 6. An adaptive protection mechanism for abnormal operating conditions is set up to monitor the deviation between the surface temperature of each section and the set value in real time. When the deviation exceeds the preset threshold, the heating power of that section is immediately forced to zero and an alarm signal is issued. Fault detection and power cut-off are completed in a short time (within 100ms), effectively preventing overheating accidents caused by sensor or actuator failures and ensuring the safety of equipment and products. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of a method for real-time temperature field compensation and intelligent control in a roll heat treatment process according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a real-time temperature field compensation and intelligent control system for a roll heat treatment process provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0021] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0022] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0023] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0025] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0027] See appendix Figure 1 An embodiment of the present invention discloses a method for real-time compensation and intelligent control of the temperature field in a roll heat treatment process, the method specifically including the following steps: S1. Establish an axial partitioned heat conduction model for the heat treatment process of the roll. Divide the roll into N equidistant control sections along the axial direction. Each section is equipped with an independent thermocouple temperature measuring point and an independent heating actuator. Construct the thermal coupling relationship matrix between each section based on Fourier's law of thermal conductivity. In specific implementation, step S1 specifically includes: S1-1: Establish the heat balance equations for each section: Fourier's law of heat conduction is the fundamental law of heat transfer, stating that the amount of heat passing through a cross-section per unit time is directly proportional to the rate of temperature change perpendicular to that cross-section and the area of ​​the cross-section, and the direction of heat transfer is opposite to the direction of temperature increase. Its mathematical expression is: , where q is the heat flux density, representing the amount of heat passing through a unit area per unit time, with units of W / m²; λ is the thermal conductivity of the material, with units of W / (m·K), a physical quantity characterizing the thermal conductivity of the material; / The temperature gradient represents the rate of change of temperature in the x-direction, with units of K / m; the negative sign indicates that the direction of heat transfer is towards the direction of decreasing temperature.

[0028] Based on Fourier's law of thermal conductivity, the specific method for constructing an axial partitioned heat conduction model for the roll is as follows: Let the thermal conductivity of the roll material be λ, its density be ρ, its specific heat capacity be c, and the temperature at the center point of the i-th partition be T. i Let Δx be the distance between the i-th segment and the (i+1)-th segment, then the heat flux density q from the i-th segment to the (i+1)-th segment is... i→i+1 For: q i→i+1 =-λ·(T i+1 -T i The formula ) / Δx represents the heat flux density between adjacent sections, which is directly proportional to the temperature difference and inversely proportional to the distance. The negative sign indicates that heat flows from the high-temperature section to the low-temperature section. The heat balance equation for each section is: The equation means that the change in heat stored in the i-th segment per unit time (left-hand side) equals the external heating power plus the heat transferred from adjacent segments minus the heat transferred out to adjacent segments. Where V i Let be the volume of the i-th segment, in m³. 3 Q i The heating power provided to the heating actuator in the i-th section, in W; A i-1 The contact area between the (i-1)th segment and the ith segment is expressed in m². 2 A i The contact area between segment i and segment i+1 is expressed in m². 2 .

[0029] S1-2. Discretize the above heat balance equation and write it in matrix form: Where C is a diagonal matrix of heat capacity, and its diagonal elements C ii =ρcV i , where represents the heat capacity of the i-th segment; K is the thermal coupling matrix, a symmetric matrix whose diagonal elements represent the heat conduction loss of the segment itself, and whose off-diagonal elements represent the thermal coupling relationship between segments; T is the temperature vector, which is composed of the surface temperatures Ti of all N segments. iThe column vector formed by T = [T1, T2, ..., T] N ] T Q is the heating power vector, which consists of the heating power Q of all N segments. i The column vector formed by Q = [Q1, Q2, ..., Q...] N ] T ; The off-diagonal element k of the thermal conductivity coupling matrix K ij The influence coefficient of heat conduction from segment j to segment i is represented by the thermal equilibrium method used to discretize Fourier's law of thermal conductivity in the one-dimensional steady-state heat conduction problem. By expressing the heat conduction coefficient between adjacent nodes as the material's thermal conductivity multiplied by the contact area divided by the node center distance, the off-diagonal element k can be obtained. ij The calculation formula is k ij =λA ij / △x ij A ij The contact area between adjacent sections, △x ij k is the center distance between adjacent segments. When |ij|>1, k ij =0 indicates that there is a direct thermal coupling relationship only between adjacent segments, and non-adjacent segments are indirectly coupled through adjacent segments.

[0030] The above-mentioned axial partitioned heat conduction model can quantitatively describe the heat transfer law along the axial direction inside the roll, providing a basic constraint for subsequent coordinated control.

[0031] S2. Construct a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculate the convective heat transfer coefficient between the roll surface and the furnace environment in real time, and dynamically correct the output value of heating power in each section according to the deviation between the measured temperature of the roll surface and the set temperature; in specific implementation, step S2 specifically includes: S2-1. Calculate the convective heat transfer q between the surface of the roll in the i-th section and the furnace environment. conv，i : Newton's law of cooling is the fundamental law describing convective heat transfer, stating that the amount of heat transferred between a surface of an object and the surrounding fluid is directly proportional to the temperature difference between the surface of the object and the fluid. Its mathematical expression is q = h(T). s -T f Where q is the convective heat flux density, in W / m²; h is the convective heat transfer coefficient, in W / (m²·K), a physical quantity characterizing the heat transfer intensity; T s The surface temperature of an object, expressed in K or °C; T f The fluid temperature is expressed in K or °C.

[0032] Based on Newton's law of cooling, the convective heat transfer q between the surface of the i-th roll section and the corresponding furnace environment is calculated.conv，i q conv，i =h i (T) s，i -T ∞，i ), where h i T is the convective heat transfer coefficient of the i-th segment, in units of W / (m²·K); s，i T represents the surface temperature of the roll in the i-th section, expressed in K or °C. ∞，i The furnace ambient temperature of the i-th section is expressed in K or °C.

[0033] The convective heat transfer coefficient h is not a constant, but a dynamic parameter influenced by various factors. This invention uses an empirical formula for natural convection heat transfer to calculate the convective heat transfer coefficient h: h = N u ·λ air / L; where N u λ is the Nusselt number, a dimensionless number representing the ratio of convective heat transfer intensity to conductive heat transfer intensity; air is the thermal conductivity of air, in W / (m·K), which can be obtained from a table; L is the characteristic dimension, which is usually the diameter of the roll for horizontally placed rolls, in meters. The Nusselt number N u The calculation uses the empirical correlation for natural convection heat transfer: N u =C·(G r ·P r ) n ; where G r The Grashof number is a dimensionless number representing the ratio of buoyancy to viscous force in natural convection; P r The Prandtl number is a dimensionless number that represents the ratio of momentum diffusion capacity to heat diffusion capacity; C and n are empirical constants. For natural convection in a horizontal cylinder, C is usually taken as 0.53 and n as 0.25 in laminar flow, and C is usually taken as 0.13 and n as 0.33 in turbulent flow. By deriving the dimensionless nature of the governing equations using Buckingham's π theorem, the formula for calculating the Grashof number Gr is: Gr = gβ(T s -T ∞ L 3 / ν; where g is the acceleration due to gravity, taken as 9.8 m / s²; β is the coefficient of thermal expansion of air, for an ideal gas, β≈1 / T ∞ The unit is K. - ¹; ν is the kinematic viscosity of air, in m² / s, which can be obtained from a table; L is the characteristic dimension (same as above), in meters; T s T represents the surface temperature. ∞ The fluid temperature.

[0034] The formula for calculating the Prandtl number Pr in fluid mechanics is: P r =ν / α; where ν is the kinematic viscosity of air, in m² / s; α is the thermal diffusivity of air, in m² / s, representing the ability of an object to achieve uniform temperature during unsteady-state heat conduction. In heat transfer analysis, the formula α = λ is derived from the definition of thermal diffusivity. air / (ρ air -c ρ，air ), where λ air air density, unit: kg / m³; c ρ，air This is the specific heat capacity of air at constant pressure, expressed in J / (kg·K).

[0035] S2-2, Based on the real-time calculated h i The value is used to dynamically correct the heating power compensation amount △Q in the i-th segment. i △Q i =h i A s，i (T) s，i -T set，i A s，i Let T be the surface area of ​​the i-th segment. set，i Set the temperature T for the surface of the roll in the i-th section. s，i Let be the measured surface temperature of the i-th segment. This formula is directly derived from Newton's law of cooling: q = h·A·(T) s -T ∞ The combined application of the principle of energy balance has the physical meaning of: to eliminate the surface temperature deviation (T) caused by convective heat transfer in the i-th section. s，i -T set，i The additional heating power compensation ΔQ required is... i It should be equal to the current convective heat transfer coefficient h of that section. i The heat exchange between the set temperature and the actual surface temperature. When the measured surface temperature is lower than the set temperature, ΔQ i If the value is positive, the heating power needs to be increased to compensate for the heat lost to the environment; when the measured surface temperature is higher than the set temperature, ΔQ i If the value is negative, the heating power needs to be reduced to avoid overheating.

[0036] S3. Establish a dynamic estimation model for the core temperature of the roll based on the lumped parameter method. Using the measured temperature of the roll surface and the time constant, calculate the dynamic trend of the core temperature in real time. When there is a deviation between the estimated core temperature and the target temperature, adjust the heating power in advance to compensate for thermal inertia hysteresis. Step S3 specifically includes: S3-1. Describe the dynamic relationship between the core temperature and surface temperature of the roll: The lumped parameter method is a simplified approach for handling unsteady heat conduction problems. When the internal thermal resistance of an object is much smaller than the surface thermal resistance, the internal temperature distribution can be approximated as uniform, with the temperature changing only with time. For high-heat-capacity components like rolling mill rolls, while the lumped parameter method doesn't fully meet all applicable conditions, a time constant can be introduced to approximate the dynamic relationship between the core and surface temperatures: Let T be the core temperature of the i-th section of the rolling mill roll. core，i (t) and its surface temperature T s，i (t) satisfies the following step response relationship: (T) s，i (t)-T ∞，i (t) / (T) core，i (t)-T ∞，i (t)) = 1 - exp(-t / τ); The physical meaning of this formula is: under step heating conditions, the relative change in surface temperature has an exponential lag relative to the relative change in core temperature, and the degree of lag is determined by the time constant τ.

[0037] The step response equation is derived from the simultaneous solution and mathematical transformation of the law of conservation of energy and Newton's law of cooling in unsteady-state heat conduction problems: First, based on the idea of ​​the lumped parameter method, a differential equation is established in which the rate of change of internal energy of the roll is equal to the surface heat transfer: Then, we introduce the excess temperature θ, where θ = TT. ∞ The equation is transformed into the standard form dθ / dt = -1 / τ·θ, where the time constant τ = ρcV / hA s Defined by the ratio of heat capacity to heat transfer capacity; solving this first-order linear homogeneous differential equation yields θ(t) = θ0e -t / τ That is, the core temperature T core The index approximates the ambient temperature, where T represents the object's temperature. ∞ Indicates ambient temperature (fluid temperature), A s This represents the surface area of ​​the roll. Based on this, by considering the phase difference between the surface temperature response and the core temperature response, the above exponential solutions are algebraically combined and transformed to finally derive the dynamic relationship between core temperature and surface temperature (T). s (t)-T ∞ ) / (T core (t)-T ∞ )=1-e -t / τ This formula accurately describes the physical law that, under step heating conditions, the relative change in surface temperature lags behind the relative change in core temperature.

[0038] Where τ is the thermal time constant of the roll, and its calculation formula is: τ=ρcV i / h i ·A s，i V iLet A be the roll volume of the i-th segment. s，i This represents the surface area of ​​the rolls in that section. This formula originates from solving the unsteady-state heat conduction differential equation using the lumped parameter method, as described above, by rearranging the energy balance equation. Furthermore, by introducing the exponential solution form, the time constant τ = ρcV / hA is naturally defined. s The time constant τ is used to characterize the magnitude of an object's thermal inertia. The physical meaning of the formula for calculating τ is: the time constant τ is equal to the heat capacity (ρcV) of the roll and the surface heat transfer capacity (hA). s The ratio of heat capacity to time constant is as follows: the greater the heat capacity, the stronger the ability of the roll to store heat, and the larger the time constant; the stronger the surface heat exchange capacity, the faster the heat exchange, and the smaller the time constant.

[0039] S3-2, Based on the measured surface temperature sequence T of the i-th segment s，i (t) k The core temperature T of the i-th segment at the current time is estimated using a reverse solution method. core，i (t) k ), T core，i (t) k )=T ∞，i +(T) s，i (t) k )-T ∞，i ) / (1-exp(-t k / τ)), where t k This represents the k-th sampling time. The derivation of this formula is based on the inverse solution of the aforementioned step response relation, using the measured surface temperature at the current time to infer the current core temperature.

[0040] S3-3, When the estimated core temperature T of the i-th segment... core，i (t) k ) and the target core temperature T of the i-th segment core，i，target When a deviation exists, calculate the advance compensation power ΔQ of the i-th segment. core，i ΔQ core，i =ρcV i / △t·(T core，i，target -T core，i (t) k )); where Δt is the control period.

[0041] This proactive compensation mechanism allows for the adjustment of heating power in advance, before the surface temperature deviates significantly, to compensate for the lag in core temperature and achieve the control effect of "measuring the surface to understand the interior and compensating in advance".

[0042] S4. The above thermal coupling relationship matrix, boundary heat transfer dynamic compensation model and core temperature estimation model are solved in a coordinated manner to generate real-time power control commands for each section, and drive the heating actuator to perform layered and zoned coordinated heating control of the rolls.

[0043] In practical implementation, the core of collaborative solution is to construct a comprehensive control objective function and, under the condition of satisfying physical constraints, solve for the optimal heating power allocation. Step S4 includes constructing the comprehensive control objective function J: ; Among them, w s,i w represents the weighting coefficient for the surface temperature deviation of the i-th segment, reflecting the importance attached to the surface temperature control accuracy of this segment; core T is the weighting coefficient for core temperature deviation, reflecting the importance attached to the accuracy of core temperature control; core，set The average core temperature estimate for the entire roll (T for each section) core，i (t) k (arithmetic mean of T) core，target The core temperature target value is uniform across the entire roll, and is a preset value.

[0044] The first term of the objective function represents the weighted sum of squares of the surface temperature deviations in each section, and the second term represents the weighted sum of squares of the core temperature deviations. By minimizing J, both surface temperature uniformity and core temperature accuracy can be simultaneously achieved.

[0045] Under the conditions of satisfying the thermal coupling relationship matrix constraint in step S1 and the boundary heat transfer constraint in step S2, solve for the heating power Q of each section that minimizes the objective function J. i This is a constrained optimization problem, which can be solved using the Lagrange multiplier method or numerical optimization algorithms. The optimal power Q obtained from the solution is... i As a real-time power control command, it is output to each heating actuator to realize the layered and zoned coordinated heating control of the rolls.

[0046] Under the conditions of satisfying the thermal coupling relationship matrix constraint in step S1 and the boundary heat transfer constraint in step S2, solve for the heating power Q of each section that minimizes the objective function J. i This is a constrained quadratic programming problem. The detailed mathematical modeling and solution process is as follows: 1. Mathematical formulation of the problem Suppose the rolling mill roll is divided into N sections along the axial direction, and the decision variable is the heating power Q of each section. i (i=1,…,N) and the surface temperature T of each section i (i.e., the section center temperature in step S1). The constraints are the steady-state heat balance equation in step S1 and the dynamically changing heat transfer coefficient h in step S2. i constitute.

[0047] 1.1 Steady-state thermal equilibrium constraint in step S1 The differential equation of step S1 is applied in steady state (dT) iDiscretize / dt), and consider the boundary convection heat transfer calculated in step S2 to obtain the energy balance equation for each section: Q i =h i ·A s，i ·(T s，i -T ∞，i )+λ·A i-1 / △x·(T i -T i-1 )-λ·A i / △x·(T i+1 -T i ), i=1,...,N. (1) Where h i The convective heat transfer coefficient of the i-th section (derived from step 2 based on the currently measured roll surface temperature T) s，i and furnace ambient temperature T ∞，i实时计算 (A is considered a known constant during optimization). s，i λ is the surface area of ​​the i-th segment; λ is the thermal conductivity of the roll material; A i Let A0 be the contact area between the i-th segment and the (i+1)-th segment (for boundary i=1, A0=0; i=N, A0=0). N =0); △x is the center distance between adjacent sections. Equation (1) can be written in matrix form: Q=GT+d (2), where T is the temperature vector, which is the temperature vector of the roll surface T of all N sections. s，i The column vector formed, T=[T s，1 T s，2 ... T s，N ] T Q is the heating power vector, representing the heating power Q of all N segments. i The column vector formed by Q = [Q1, Q2, ..., Q...] N ] T G is the thermal conductivity coupling matrix, which is an N×N tridiagonal matrix: G ii =h i ·A s，i +λ / △x·(A i-1 +A i ),G i，i-1 =-λ·A i-1 / △x,G i，i+1 =-λA i / △x. G ii The diagonal element represents the surface temperature T of the roll in the i-th section. s，i The power Q required by itself i G's contribution; i，i-1 The second diagonal element represents the temperature T of the adjacent section on the left. i-1 The required power Q for the i-th segment i G's contribution;i，i+1 The second diagonal element represents the temperature T of the adjacent section on the right. i+1 The required power Q for the i-th segment i The contribution; vector d is a constant vector, independent of temperature T, representing a fixed heat load caused by ambient temperature; the i-th component of d is -h i A s，i T ∞，i This part represents even if the surface temperature T of the roll is... s，i When it is zero, in order to withstand the ambient temperature T ∞，i The resulting heat loss (or gain) theoretically requires an additional (or reduced) reference power.

[0048] 1.2 Objective Function The objective function J consists of the surface temperature deviation and the core temperature deviation: (3); Among them, w s,i w represents the weighting coefficient for the surface temperature deviation of the i-th segment, reflecting the importance attached to the surface temperature control accuracy of this segment; core T is the weighting coefficient for core temperature deviation, reflecting the importance attached to the accuracy of core temperature control; core，set T is the estimated average core temperature of the entire roll. core，target The target core temperature is set uniformly for the entire roll. Based on the lumped parameter model in step S3, it is assumed that the entire roll has a uniform core temperature, and the average surface temperature of all sections is taken. Take the average value of the furnace ambient temperature in all sections. Then T core，set =T ∞，avg +(T) s，avg -T ∞，avg ) / (1-e -t / τ Where t is the current heat treatment time; τ is the thermal time constant of the entire roll, τ=ρcV / h avg ·A s h avg The average convective heat transfer coefficient is calculated based on average heat transfer conditions. A s T is the surface area of ​​the entire roll; V is the volume of the entire roll. The above formula... core，set =T ∞，avg +(T) s，avg -T ∞，avg ) / (1-e -t / τ It can be linearized to T core，set =c T T+d0(4), where c is an N-dimensional vector, and each component c i =1 / (N(1-e-t / τ ) (τ is the thermal time constant of the i-th segment), representing the surface temperature T of the i-th segment. s，i Estimate the average core temperature T of the entire roll core，set Contribution weights; constant d0=T ∞，avg ·(1-1 / (1-e -t / τ (τ is the thermal time constant of the entire roll), representing the average value T of the furnace ambient temperature. ∞，avg The average core temperature T of the entire roll core，set Estimated fixed contribution.

[0049] Substitute (4) into (3) and write it as a quadratic form: J(T) = (TT) set ) T ·W s · (TT) set )+w core ·(c T T+d0-T core，target ) 2 W s =diag(w s，1 , ..., w s，N (5) Where J(T) represents the objective function with temperature vector T as the independent variable, and its value quantitatively evaluates the degree of conformity between the current temperature distribution and the ideal process requirements; T is the measured surface temperature vector of each section; T set A vector of surface temperature settings for each section; W s This is a surface temperature weighted diagonal matrix, reflecting the degree of importance attached to the control accuracy of each section; the diagonal elements w s，i This indicates the level of importance attached to the surface temperature control accuracy of the i-th segment; (TT) set ) T ·W s · (TT) set ) represents the weighted sum of squares of surface temperature deviations to ensure axial uniformity; c is an N×1 contribution weight vector that converts the surface temperature into a core temperature estimate, with each component c i =1 / (N(1-e -t / τ ) represents the contribution of the surface temperature of the i-th segment to the average core temperature estimate; d0 is the ambient temperature compensation constant; c T T+d0 is the estimated average core temperature; w core ·(c T T+d0-T core，target ) 2 This is a weighted square of the core temperature deviation to ensure sufficient core microstructure transformation.

[0050] 1.3 Power Upper and Lower Limit Constraints Actual heaters have power limitations: Q min，i ≤Q i ≤Q max，i , i=1,...,N; (6) 2. Optimize problem transformation Since Q and T are linearly related through equation (2), we can choose to use Q or T as the decision variable. To facilitate handling inequality constraints, we usually use Q as the variable and express T as a function of Q.

[0051] From (2), we can obtain (assuming G is invertible): T = G -1 (Qd) (7) Substituting equation (7) into equation (5), we obtain a quadratic function of Q: J(Q) = (G) -1 (Qd)-T SET ) T ·W s ·(G) -1 (Qd)-T SET )+w core ·(c T ·G -1 (Qd) + d0 - T core，target ) 2 After unfolding and rearranging, it becomes a standard quadratic form: J(Q) = 1 / 2Q T ·H·Q+f T ·Q+const (8) Where H is a symmetric positive definite matrix (if the weights are appropriate), reflecting the coupling relationship between the power of each segment—changing the power of a segment not only affects the temperature of that segment, but also affects adjacent segments through heat conduction, thus affecting the entire objective function. f is the gradient vector, reflecting the driving force of the deviation between the current setpoint and the measured value on power adjustment. If the temperature of a certain segment deviates significantly from the setpoint, the corresponding component in f will be larger, driving the optimization algorithm to increase or decrease the power of that segment. The specific expression can be obtained by matrix operations. const represents a constant term, representing the "inherent deviation" that cannot be eliminated even under ideal power, such as the deviation caused by the ambient temperature T. ∞，avg The resulting fixed heat load. This part does not affect the optimal power allocation, but only the absolute value of the objective function.

[0052] Therefore, the original problem is transformed into a quadratic programming problem: (9) 2. Solution Method Analytical solution without inequality constraints (Lagrange multiplier method) If the upper and lower limits of power are ignored, T can be solved directly. By introducing the Lagrange multiplier vector μ, the equality constraint (2) Q=GT+d is incorporated into the objective, and the Lagrange function is constructed. L(T, Q, μ) = J(T) + μ T ·(QG·Td) Find the partial derivative and set it to zero: From the second equation, we get μ=0. Substituting this into the first equation, we get: (W) s +w core ·c·c T )·T=W s ·T set -w core ·c·(d0-T core，target ) Solving this system of linear equations yields the optimal T. Then, from (2), we get Q. =G·T +d, this is the optimal control command under unconstrained conditions, Q It is a vector containing N components, each component vector Q. i The heating power corresponding to different sections i.

[0053] In real-time control, the following process is executed in each control cycle: Data acquisition: Read the surface temperature T of each zone. i Ambient temperature T ∞ The current core temperature T is estimated by step S3. core，est； Calculate the heat transfer coefficient: According to the formula in step S2, using the current T i T ∞ Calculate the convective heat transfer coefficient h for each zone. i ; Constructing matrices: Assemble matrices G and d, as well as H and f in the objective function (based on the current time t and weights); Optimization solution: Call the quadratic programming solver to obtain the optimal power Q. ; Command output: Q Each component Q i The corresponding heating actuators in each section are sent; Update status: Wait for the next sampling time and repeat the above process.

[0054] The above method enables real-time coordinated control of the temperature field of the roll in a layered and zoned manner, minimizing the temperature deviation between the surface and the core while meeting physical constraints, thereby improving the quality of heat treatment.

[0055] In practice, the value of N ranges from 3 to 20. Each section has an independently configured thermocouple temperature measuring point located on the surface of the roll at the center of that section. Each section also has an independently configured heating actuator, either an induction heating coil or a resistance heating element. When N < 3, the axial partitioning is too coarse, failing to effectively address the temperature difference between the ends and the middle of the roll. When N > 20, the complexity of the control system increases dramatically, and the thermal coupling relationship between adjacent sections becomes overly complex, potentially reducing control efficiency.

[0056] In practice, the furnace ambient temperature T in the i-th section is... ∞，i The method of obtaining the temperature is as follows: an independent ambient temperature sensor is set in the furnace space corresponding to each control section i to collect the local ambient temperature of the section in real time, which is used for the accurate calculation of the boundary heat transfer compensation model of the section.

[0057] In practice, the thermal time constant τ of the roll is dynamically updated using a segmented identification method: the heat treatment process is divided into a heating stage, a holding stage, and a cooling stage, and the thermal time constant τ of each stage is identified separately. heat τ hold and τ cool This is used for core temperature estimation at each stage. The thermal time constant τ of the roll is dynamically updated using a segmented identification method, aiming to address the physical nature of the non-constant thermal time constant: thermal time constant τ = ρcV / hA s In the equation, ρ, c, and V all change with temperature (material thermal properties change with temperature, and the convective heat transfer coefficient also changes with temperature). Therefore, τ is not constant. During the heating phase: the temperature difference is large, convection is strong, h is large, τ is small, and the thermal response is fast. During the holding phase: the temperature difference is small, convection weakens, h is small, τ is large, and the thermal response is slow. During the cooling phase: similar to the heating phase but in the opposite direction. Substitute the τ value for the corresponding phase into T. core，i (t) k )=T ∞ +(T) s，i (t) k )-T ∞ ) / (1-exp(-t k / τ)), the estimated result is closer to the true value.

[0058] The weighting coefficient w s，i and w core，i A dynamic adjustment strategy is adopted: during the heating stage, the weight of the core temperature deviation w is increased. core，i To accelerate core heating; during the heat preservation stage, increase the weighting of surface temperature deviation w. s，iTo ensure surface temperature uniformity, during the cooling phase, the weights of both factors are balanced to achieve uniform cooling. The weighting coefficients are dynamically adjusted to adapt to the core control objectives of different process stages. During the heating phase, rapid and uniform heating is achieved by increasing w. core Prioritize ensuring the core temperature keeps up to prevent surface overheating while the core remains sluggish; during the heat preservation stage, ensure uniform and stable temperature, and increase w s，i Prioritize ensuring uniform temperature across all surface regions to guarantee consistent microstructure transformation; during the cooling phase, control thermal stress and balance both to prevent excessive temperature differences that could lead to thermal stress cracking. A single fixed weight cannot simultaneously meet the control requirements of different stages.

[0059] It also includes step S5, the abnormal operating condition adaptive protection step: when the measured surface temperature T of any section s，i With the set temperature T set，i The absolute value of the deviation exceeds the preset threshold ΔT max When this occurs, the abnormal protection mechanism is triggered, reducing the heating power Q of this section. i Forced zeroing and alarm signal. In abnormal situations such as temperature runaway or sensor failure, the heating power is forcibly cut off and an alarm is triggered to prevent the rolls from overheating and being damaged, avoid the generation of scrap, and ensure the safety of equipment and operation. It is a safety barrier for the reliable operation of the control system.

[0060] The present invention will be further described in detail below with reference to specific embodiments, but the scope of protection of the present invention is not limited thereto.

[0061] Examples are given below: This embodiment uses the quenching treatment of a large roll with a diameter of 500mm and a length of 3000mm as an example to illustrate the method of the present invention in detail.

[0062] Step 1: Axial partitioning The roll is divided into six equidistant control sections along the axial direction, each section being 500 mm long. A type K thermocouple is installed on the roll surface at the center of each section for real-time surface temperature monitoring. Each section is equipped with an independent induction heating coil as the heating actuator, with a coil width of 400 mm covering the main heating area of ​​that section. Thermal insulation rings are installed between the sections to reduce direct heat radiation interference without blocking heat conduction.

[0063] The roll material is Cr5 alloy tool steel, with the following thermophysical properties: thermal conductivity λ = 35 W / (m·K), density ρ = 7800 kg / m³, and specific heat capacity c = 460 J / (kg·K). Calculate the volume V of each section. i =π×(0.25)²×0.5≈0.098m³, heat capacity ρcV i ≈7800×460×0.098≈351,624 J / K. Contact area A between adjacent sections. i=π×(0.25)²≈0.196 m².

[0064] Based on the formula in step S1, construct the thermal conductivity coupling matrix K. Taking the 3rd and 4th segments as examples, the thermal conductivity coupling coefficient k between them is... 34 =λA 34 / Δx 34 =35×0.196 / 0.5≈13.72W / K. The same applies to other adjacent sections; the coupling coefficient for non-adjacent sections is 0.

[0065] Step 2: Boundary Heat Transfer Compensation Independent ambient temperature sensors are installed at corresponding locations in each section of the furnace to collect real-time temperature data. ∞,i Assume the measured surface temperature T of the current third section is... s,3 =850℃, set temperature T set,3 =860℃, ambient temperature T ∞,3 =820℃.

[0066] Calculate the Grashof number Gr. Take the kinematic viscosity of air ν = 1.6 × 10⁻⁶. -5 m² / s, coefficient of thermal expansion β = 1 / (820 + 273) ≈ 9.15 × 10⁻⁶ m² / s -4 K -1 Given that the characteristic dimension L = 0.5m (roll diameter), then: Gr=gβ(T s -T ∞ L³ / ν² = 9.8 × 9.15 × 10 -4 ×(850-820)×0.5³ / (1.6×10 -5 )² =9.8×9.15×10 -4 ×30×0.125 / (2.56×10 -10 ) =9.8×9.15×10 -4 ×30×0.125×3.91×10 9 ≈1.32×10 8 ; Calculate the Prandtl number Pr. Take the air thermal diffusivity α = 2.2 × 10⁻⁶. -5 If m² / s, then Pr = ν / α = 1.6 × 10⁻⁶ -5 / 2.2×10 -5 ≈0.73.

[0067] Calculate the Nusselt number Nu. Assuming laminar flow, take C = 0.53 and n = 0.25, then: Nu = 0.53 × (Gr·Pr) 0.25 =0.53×(1.32×108 ×0.73) 0.25 =0.53×(9.64×10 7 ) 0.25 =0.53×55.6≈29.5; Calculate the convective heat transfer coefficient h. Taking the thermal conductivity of air λ_air = 0.062 W / (m·K), then: h = Nu·λ_air / L = 29.5 × 0.062 / 0.5 ≈ 3.66 W / (m²·K); Calculate the surface area A of section 3. s,3 =π×0.5×0.5≈0.785m².

[0068] Calculate the power compensation amount ΔQ3=h3A s,3 (T s,3 -T set,3 =3.66 × 0.785 × (850 - 860) ≈ -28.7W. A negative value indicates that the heating power needs to be reduced by 28.7W.

[0069] Step 3: Core Temperature Estimation Calculate the total volume of the rolls: V = π × (0.25)² × 3 ≈ 0.589 m³, and the total surface area: A. s =π×0.5×3+2×π×(0.25)²≈4.71+0.393≈5.103m².

[0070] Calculate the average convective heat transfer coefficient h avg For each section h i Take the average, assuming the average value is 3.5 W / (m²·K).

[0071] Calculate the thermal time constant τ = ρcV / (h) avg A s =7800×460×0.589 / (3.5×5.103)=2,113,092 / (17.86)≈118,300s≈32.9h. This time constant is relatively large, which is consistent with the characteristic of large thermal inertia of large rolling mill rolls.

[0072] Assume the current time t k =5400s (1.5h), measured surface temperature T s (t k )=850℃, ambient temperature T ∞ =820℃, then estimate the core temperature: T core (t k )=820+(850-820) / [1-exp(-5400 / 118300)] =820+30 / [1-exp(-0.0456)] =820+30 / [1-0.9554] =820 + 30 / 0.0446 ≈ 820 + 672 ≈ 1492 K ≈ 1219 ℃ This estimate indicates that the core temperature is still much lower than the surface temperature, which is consistent with the actual situation.

[0073] If the target core temperature T core,target If the core temperature is 860℃, then the core temperature deviation is 1219-860=359℃, requiring a reduction in core temperature. However, in reality, roll quenching requires rapid heating, and the estimated core temperature here is too high, indicating that the τ value needs adjustment or segmented identification needs to be adopted.

[0074] Adjust the τ value to perform segmented identification. Based on experience, during the heating stage, τ... heat Take 80,000 seconds. Recalculate: T core (t k = 820 + 30 / [1 - exp(-5400 / 80000)] =820+30 / [1-exp(-0.0675)] =820+30 / [1-0.9347] =820 + 30 / 0.0653 ≈ 820 + 459 ≈ 1279 K ≈ 1006 ℃ This estimate is more reasonable. Calculate the lead compensation power ΔQ. core =ρcV / Δt×(T core,target -T core (t k Taking Δt = 60s, then: ΔQ core =7800×460×0.589 / 60×(860-1006) =2,113,092 / 60×(-146) =35,218×(-146)≈-5,141,828W≈-5.14MW A negative value indicates that the heating power needs to be significantly reduced to slow down the core temperature rise. This value is relatively large and needs to be limited in conjunction with other constraints during actual implementation.

[0075] Step 4: Cooperative Solving Construct the objective function J, and take w s,i =1 (i=1~6), w core =10 (Emphasis on core control during the heating phase). Set the target core temperature T. core,target =860℃, surface set temperature T set,i =860℃.

[0076] Under the conditions of thermal coupling constraint in step S1 and boundary heat transfer constraint in step S2, solve for the optimal power allocation Q. i Due to the complexity of the calculations, this embodiment simplifies it to: calculating ΔQ in step S2... i And ΔQ calculated in step S3 core The data is allocated proportionally to each section, while also taking into account the thermal coupling effects between sections.

[0077] The final generated control command is as follows: the heating power of sections 1 and 6 (ends) increases by 5%, the heating power of sections 2-5 (middle) decreases by 3%, and the overall heating power is adjusted according to ΔQ. core The scaling adjustment is performed. This command is output to each induction heating power supply to execute coordinated heating control.

[0078] As another embodiment of the present invention, see Appendix Figure 2 A real-time temperature field compensation and intelligent control system for a roll heat treatment process is disclosed, comprising a first module, a second module, a third module and a fourth module, which are connected in sequence. The first module is used to establish an axial partitioned heat conduction model for the heat treatment process of the roll, dividing the roll into N equidistant control sections along the axial direction. Each section is equipped with an independent thermocouple temperature measuring point and an independent heating actuator. Based on Fourier's law of thermal conductivity, a thermal coupling relationship matrix between each section is constructed. The second module is used to construct a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculate the convective heat transfer coefficient between the roll surface and the furnace environment in real time, and dynamically correct the output value of heating power in each section according to the deviation between the measured temperature of the roll surface and the set temperature. The third module is used to establish a dynamic estimation model of roll core temperature based on the lumped parameter method. It uses the measured temperature of the roll surface and the time constant to calculate the dynamic change trend of the roll core temperature in real time. When there is a deviation between the estimated core temperature and the target temperature, the heating power is adjusted in advance to compensate for thermal inertia hysteresis. The fourth module is used to collaboratively solve the above-mentioned thermal coupling relationship matrix, boundary heat transfer dynamic compensation model and core temperature estimation model, generate real-time power control commands for each section, and drive the heating actuator to perform layered and zoned collaborative heating control of the rolls.

[0079] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0080] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0081] See appendix Figure 3 Another embodiment of the present invention discloses a computer device, including: a memory and a processor and a computer program stored in the memory, wherein when the computer program is executed on the processor, a method for real-time compensation and intelligent control of the temperature field in a roll heat treatment process as described in any of the above claims is implemented.

[0082] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for real-time temperature field compensation and intelligent control of a roll heat treatment process as described in any of the preceding embodiments.

[0083] In the embodiments disclosed in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0084] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

Claims

1. A method for real-time temperature field compensation and intelligent control in a roll heat treatment process, characterized in that, The method specifically includes the following steps: S1. Establish an axial partitioned heat conduction model for the heat treatment process of the roll. Divide the roll into N equidistant control sections along the axial direction. Each section is equipped with an independent thermocouple temperature measuring point and an independent heating actuator. Construct the thermal coupling relationship matrix between each section based on Fourier's law of thermal conductivity. S2. Construct a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculate the convective heat transfer coefficient between the roll surface and the furnace environment in real time, and dynamically correct the output value of heating power in each section according to the deviation between the measured temperature of the roll surface and the set temperature. S3. Establish a dynamic estimation model for the core temperature of the roll based on the lumped parameter method. Utilize the measured temperature of the roll surface and the time constant to calculate the dynamic change trend of the core temperature of the roll in real time. When there is a deviation between the estimated core temperature and the target temperature, adjust the heating power in advance to compensate for thermal inertia hysteresis. S4. The above thermal coupling relationship matrix, boundary heat transfer dynamic compensation model and core temperature estimation model are solved in a coordinated manner to generate real-time power control commands for each section, and drive the heating actuator to perform layered and zoned coordinated heating control of the rolls.

2. The method for real-time temperature field compensation and intelligent control in a roll heat treatment process as described in claim 1, characterized in that, Step S1 specifically includes: S1-1. Establish the heat balance equations for each section: Let the thermal conductivity of the roll material be λ, the density be ρ, the specific heat capacity be c, and the temperature at the center point of the i-th section be T. i Let Δx be the distance between the center of segment i and the center of segment i+1. Then the heat flux density q from segment i to segment i+1 is... i→i+1 For: q i→i+1 =-λ·(T i+1 -T i The heat balance equations for each section are: ) / Δx V i Let Q be the volume of the i-th segment. i The heating power A provided to the heating actuator of the i-th section i-1 Let A be the contact area between the (i-1)th segment and the ith segment. i Let be the contact area between the i-th segment and the (i+1)-th segment; S1-2. Discretize the above heat balance equation and write it in matrix form: Where C is the heat capacity diagonal matrix, K is the thermal conductivity coupling matrix, T is the temperature vector, and Q is the heating power vector; the off-diagonal elements k of the thermal conductivity coupling matrix K are... ij The coefficient representing the influence of segment j on the heat conduction of segment i is calculated using the formula k. ij =λA ij / △x ij A ij The contact area between adjacent sections, △x ij This represents the center-to-center distance between adjacent segments.

3. The method for real-time temperature field compensation and intelligent control in a roll heat treatment process as described in claim 1, characterized in that, Step S2 specifically includes: S2-1. Calculate the convective heat transfer q between the surface of the roll in the i-th section and the furnace environment. conv，i q conv，i =h i (T) s，i -T ∞，i ), where h i Let T be the convective heat transfer coefficient of the i-th segment. s，i Let T be the surface temperature of the roll in the i-th section. ∞，i Let be the furnace ambient temperature of the i-th section; S2-2, Based on the real-time calculated h i The value is used to dynamically correct the heating power compensation amount △Q in the i-th segment. i △Q i =h i A s，i (T) s，i -T set，i ), where A s，i Let T be the surface area of ​​the i-th segment. set，i Set the temperature for the surface of the roll in the i-th section.

4. The method for real-time temperature field compensation and intelligent control in the heat treatment process of rolls as described in claim 3, characterized in that, The convective heat transfer coefficient h is calculated using the empirical formula: h = N u ·λ air / L; where N u The Nusselt number is a dimensionless number that represents the ratio of convective heat transfer intensity to thermal conductivity intensity. λ air is the thermal conductivity of air, in W / (m·K), which can be obtained from a table; L is the characteristic dimension, which is usually the diameter of the roll for horizontally placed rolls, in meters. The Nusselt number N u The calculation uses the empirical correlation for natural convection heat transfer: N u =C·(G r ·P r ) n ; where G r The Grashof number is a dimensionless number representing the ratio of buoyancy to viscous force in natural convection; P r The Prandtl number is a dimensionless number that represents the ratio of momentum diffusion capacity to heat diffusion capacity; C and n are empirical constants. For natural convection in a horizontal cylinder, C is usually taken as 0.53 and n as 0.25 in laminar flow, and C is usually taken as 0.13 and n as 0.33 in turbulent flow. The formula for calculating the Grashof number Gr is: Gr = gβ(T) s -T ∞ L 3 / ν; where g is the acceleration due to gravity, taken as 9.8 m / s²; β is the coefficient of thermal expansion of air, for an ideal gas, β≈1 / T ∞ The unit is K. -1 ν is the kinematic viscosity of air, in m² / s, which can be obtained from a table; L is the characteristic dimension (same as above), in meters. The formula for calculating Prandtl's number Pr is: P r =ν / α; where α is the thermal diffusivity of air, in m² / s, α=λ air / (ρ air -c ρ，air ), where λ air For air density, c ρ，air This refers to the specific heat capacity of air at constant pressure.

5. The method for real-time temperature field compensation and intelligent control in a roll heat treatment process as described in claim 3, characterized in that, Step S3 specifically includes: S3-1. Let the core temperature of the roll in the i-th section be T. core，i (t) and its surface temperature T s，i (t) satisfies the following relation: (T) s，i (t)-T ∞，i ) / (T core，i (t)-T ∞，i ) = 1 - exp(-t / τ); where τ is the thermal time constant of the roll, and its calculation formula is: τ = ρcV i / h i A s，i V i Let A be the volume of the i-th section of the roll. s，i Let be the surface area of ​​the i-th section of the roll; S3-2, Based on the measured surface temperature sequence T of the i-th segment s，i (t) k The core temperature T of the i-th segment at the current time is estimated using a reverse solution method. core，i (t) k ), T core，i (t) k )=T ∞，i +(T) s，i (t) k )-T ∞，i ) / (1-exp(-t k / τ)), where t k This represents the k-th sampling time. S3-3, When the estimated core temperature T of the i-th segment... core，i (t) k ) and the target core temperature T of the i-th segment core，i，target When a deviation exists, calculate the advance compensation power ΔQ of the i-th segment. core，i ΔQ core，i =ρcV i / △t·(T core，i，target -T core，i (t) k )); where Δt is the control period.

6. The method for real-time temperature field compensation and intelligent control in a roll heat treatment process as described in claim 3, characterized in that, Step S4 includes constructing the comprehensive control objective function J: ； Among them, w s,i w represents the weighting coefficient for the surface temperature deviation of the i-th segment, reflecting the importance attached to the surface temperature control accuracy of this segment; core T is the weighting coefficient for core temperature deviation, reflecting the importance attached to the accuracy of core temperature control; core，set The average core temperature estimate for the entire roll (T for each section) core，i (t) k (arithmetic mean of T) core，target The target core temperature is set uniformly for the entire roll.

7. The method for real-time temperature field compensation and intelligent control in a roll heat treatment process as described in claim 1, characterized in that, The value of N ranges from 3 to 20. The thermocouple temperature measuring point of each section is located on the surface of the roll at the center of the section. The heating actuator of each section is an induction heating coil or a resistance heating element. The furnace ambient temperature T in section i ∞，i The method of obtaining the temperature is as follows: an independent ambient temperature sensor is set in the furnace space corresponding to each control section i to collect the local ambient temperature of the section in real time, which is used for the accurate calculation of the boundary heat transfer compensation model of the section.

8. The method for real-time temperature field compensation and intelligent control in the heat treatment process of rolls as described in claim 3, characterized in that, The thermal time constant τ of the roll is dynamically updated using a segmented identification method: the heat treatment process is divided into a heating stage, a holding stage, and a cooling stage, and the thermal time constant τ of each stage is identified separately. heat τ hold and τ cool Used for core temperature estimation at each stage; The weighting coefficient w s，i and w core，i A dynamic adjustment strategy is adopted: during the heating stage, the weight of the core temperature deviation w is increased. core，i To accelerate core heating; During the heat preservation stage, the weight of surface temperature deviation w is increased. s，i To ensure uniform surface temperature; During the cooling phase, the weights of the two are balanced to achieve uniform cooling.

9. The method for real-time temperature field compensation and intelligent control in a roll heat treatment process as described in claim 1, characterized in that, It also includes S5, an adaptive protection step for abnormal operating conditions: when the measured surface temperature T of any section... s，i With the set temperature T set，i The absolute value of the deviation exceeds the preset threshold ΔT max When this occurs, the abnormal protection mechanism is triggered, reducing the heating power Q of this section. i Force the value to zero and issue an alarm signal.

10. A real-time temperature field compensation and intelligent control system for a roll heat treatment process, characterized in that, It includes a first module, a second module, a third module, and a fourth module, which are connected in sequence; The first module is used to establish an axial partitioned heat conduction model for the heat treatment process of the roll, dividing the roll into N equidistant control sections along the axial direction. Each section is equipped with an independent thermocouple temperature measuring point and an independent heating actuator. Based on Fourier's law of thermal conductivity, a thermal coupling relationship matrix between each section is constructed. The second module is used to construct a boundary heat transfer dynamic compensation model based on Newton's law of cooling, calculate the convective heat transfer coefficient between the roll surface and the furnace environment in real time, and dynamically correct the output value of heating power in each section according to the deviation between the measured temperature of the roll surface and the set temperature. The third module is used to establish a dynamic estimation model of roll core temperature based on the lumped parameter method. It uses the measured temperature of the roll surface and the time constant to calculate the dynamic change trend of the roll core temperature in real time. When there is a deviation between the estimated core temperature and the target temperature, the heating power is adjusted in advance to compensate for thermal inertia hysteresis. The fourth module is used to collaboratively solve the above-mentioned thermal coupling relationship matrix, boundary heat transfer dynamic compensation model and core temperature estimation model, generate real-time power control commands for each section, and drive the heating actuator to perform layered and zoned collaborative heating control of the rolls.

Images