Method for optimizing process parameters of seamless tube piercer based on temperature field prediction

Abstract

This invention relates to the field of steel pipe processing technology and discloses a method for optimizing process parameters of a seamless steel pipe piercing machine based on temperature field prediction. The method includes obtaining the metal rheological velocity field of the billet and extracting the strain rate tensor; generating a second-order anisotropic thermal conductivity tensor based on the strain rate tensor, and solving for a three-dimensional temperature scalar field and a three-dimensional thermal gradient vector field; calculating the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field to obtain a local divergence penalty function; when the local divergence penalty function exceeds the limit, projecting the abnormal nodes inversely onto the target tool surface to trigger an independent high-pressure cooling mechanism to deflect the direction of the thermal gradient vector field; constructing a system-level optimization objective functional by volume integration of the local divergence penalty function, and performing nonlinear optimization iteration under physical constraints. This invention reconstructs the anisotropic thermal conductivity model, improves the accuracy of temperature field solution, and effectively suppresses the initiation of microcracks inside the billet through thermal boundary feedforward control and macroscopic process optimization.

Description

Technical Field

[0001] This invention relates to the field of steel pipe processing technology, specifically to a method for optimizing process parameters of a seamless steel pipe piercing machine based on temperature field prediction. Background Technology

[0002] The piercing process is a core hot deformation step in the production of seamless steel pipes. The piercing machine uses high-speed rotating rolls and a mandrel to forcefully pierce and deform a solid billet. Within the main deformation zone, the billet is simultaneously subjected to intense plastic shear and complex heat conduction. To ensure the internal forming quality of the billet and prevent the initiation of microcracks, engineers use computer numerical simulation to establish thermo-mechanical coupled partial differential equations for the piercing process. They solve these equations to calculate the three-dimensional temperature field distribution inside the billet and then use the temperature field prediction results to perform reverse optimization of the operating process parameters of the seamless steel pipe piercing machine.

[0003] In existing technologies, when constructing the three-dimensional steady-state thermo-mechanical coupling heat conduction control equations for tube blanks, the basic physical property test data of the tube blank material are extracted as the input boundary of the control equations. The basic physical property test data includes the tube blank material density, constant-pressure specific heat capacity, and initial isotropic thermal conductivity measured under static or extremely low strain rates. The numerical solution program substitutes the initial isotropic thermal conductivity as a fixed scalar parameter into the Hamiltonian operator, and combines the heat generation rate of the volumetric heat source converted from plastic deformation work with the heat generation rate of the frictional heat source at the tool interface. A numerical algorithm is used to discretize and solve the partial differential equations, outputting the temperature values ​​and thermal gradient vectors of each discrete node within the main deformation zone, providing a temperature boundary reference for adjusting macroscopic rolling process parameters.

[0004] Existing technologies for temperature field prediction suffer from a disconnect between the underlying physical property characterization mechanism and the extreme plastic deformation state. The piercing process of seamless steel pipes is accompanied by extreme metal rheological behavior, with intense elongation and lattice distortion of the grains within the billet along the main deformation direction. This grain elongation and lattice distortion alter the scattering paths of phonons and electrons within the metal, resulting in a significant difference in the heat transfer capacity of the billet metal parallel and perpendicular to the plastic flow direction. Current numerical calculation models use isotropic scalar thermal conductivity for heat transfer calculations, neglecting the deformation-induced anisotropic heat transfer mechanism. This leads to the constructed thermo-mechanical coupling equations failing to accurately reflect the spatial heat transfer differences caused by microstructural distortion within the main deformation zone. Ultimately, the resulting three-dimensional temperature scalar field and thermal gradient vector field deviate from the actual physical evolution state, depriving subsequent optimization of process parameters based on the biased temperature field data of accurate underlying data support. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for optimizing process parameters of seamless steel pipe piercing machines based on temperature field prediction, thus solving the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of this invention provides a method for optimizing process parameters of a seamless steel tube piercing mill based on temperature field prediction, comprising the following steps: Obtain the basic geometric parameters of the seamless steel pipe piercing machine, the initial thermophysical parameters of the billet material, and the preset process optimization variable space.

[0007] Within the process optimization variable space, a set of process parameters are selected. Based on the selected process parameters and basic geometric parameters, the metal rheological velocity vector field of discrete nodes in the main deformation zone is calculated, and the strain rate tensor is extracted from the metal rheological velocity vector field.

[0008] A second-order anisotropic thermal conductivity tensor reflecting the lattice distortion effect is generated based on the strain rate tensor. The second-order anisotropic thermal conductivity tensor is substituted into the thermo-mechanical coupling partial differential equation to obtain the three-dimensional temperature scalar field of the main deformation region. The three-dimensional thermal gradient vector field is obtained by spatial differentiation of the three-dimensional temperature scalar field.

[0009] The dot product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field is calculated to obtain the local divergence penalty function.

[0010] The local divergence penalty function is monitored. When the local divergence penalty function is greater than the preset critical threshold, the three-dimensional spatial coordinates of the abnormal node are extracted. The abnormal node is projected onto the surface of the target tool according to the normal shortest distance mapping rule to obtain the surface mapping coordinates. Based on the surface mapping coordinates, a multi-segment independent high-pressure cooling mechanism is triggered.

[0011] The volume integral of the local divergence penalty function in the main deformation zone is used to obtain the system-level optimization objective functional. The material peak temperature constraint and the equipment limit torque constraint are established by combining the billet material properties and the equipment bearing parameters.

[0012] A nonlinear programming algorithm is used to iteratively optimize the system-level objective functional under the constraints of the equipment's limit torque and the material's peak temperature. When the absolute value of the change of the system-level objective functional in adjacent iteration steps is less than or equal to the preset convergence tolerance, the globally optimal combination of process parameters is output. The globally optimal combination of process parameters is sent to the programmable logic controller to adjust the speed of the main drive shaft motor and the stroke of the pressing hydraulic cylinder of the seamless steel pipe piercing machine.

[0013] Further elaborating on the innovative physical mechanism of this invention, in conventional piercing mill temperature field solution models, the thermal conductivity properties of materials are often set as isotropic scalar parameters. In the intense plastic deformation zone of seamless steel pipe piercing, the grains inside the billet elongate along the main deformation direction of the metal, accompanied by lattice distortion, resulting in a significant difference between the thermal conductivity parallel to the plastic flow direction and the thermal conductivity perpendicular to this direction. This invention extracts the strain rate tensor from the metal rheological velocity vector field, and reconstructs the thermal properties of the billet material using the double dot product operation of the strain rate tensor and the fourth-order deformation thermal conductivity coupling tensor, generating a second-order anisotropic thermal conductivity tensor. A three-dimensional orthogonal rotation transformation matrix is ​​constructed using the eigenvectors of the strain rate tensor, maintaining geometric alignment between the principal axis of the second-order anisotropic thermal conductivity tensor and the corresponding metal main deformation direction in space. This eliminates the physical mechanism distortion phenomenon of the scalar thermal conductivity model under microstructural distortion conditions, obtaining a more accurate three-dimensional temperature scalar field and a three-dimensional thermal gradient vector field.

[0014] To further elucidate the topological mapping and control mechanism of this invention, the initiation of microcracks inside the tube blank is not only affected by the local temperature value, but also depends on the matching relationship between the direction of metal plastic deformation flow and the spatial temperature gradient. This invention constructs a local divergence penalty function by calculating the absolute value of the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field. The value of the inner product is equivalent to the convective derivative of the spatial nodal temperature, used to physically characterize the degree of dispersion of the metal rheological velocity vector from the tangential direction of the isothermal surface. The larger the local divergence penalty function, the more significant the characterization of the metal's behavior across regions of severe temperature differences, and the higher the risk of microcracks induced by additional thermal stress. This invention, on the one hand, performs three-dimensional spatial volume integral operations on the local divergence penalty function within the main deformation zone, reducing and aggregating it from distributed field data into a system-level optimization objective functional with process parameters as independent variables, and incorporating it into nonlinear optimization iteration; on the other hand, it establishes a feedforward compensation mechanism. When the local divergence penalty function exceeds the limit, it projects the internal abnormal discrete nodes inversely along the normal direction to the external target tool surface. By establishing a local asymmetric strong cooling boundary under specific spatial coordinates, it applies a directional heat flow suction effect to the inside of the tube blank, changes the local heat dissipation rate of the internal discrete nodes, and thus forcibly deflects the spatial direction of the three-dimensional thermal gradient vector field inside the tube blank, making it spatially aligned with the metal rheological velocity vector field, cutting off the conditions for thermal stress concentration and defect initiation from the bottom physical boundary.

[0015] A second aspect of the present invention provides a seamless steel tube piercing mill process parameter optimization system based on temperature field prediction, used to execute the optimization method described in the first aspect, the system comprising: At least one processor; and a memory communicatively connected to at least one processor.

[0016] The memory stores computer program instructions executable by at least one processor. When executed by at least one processor, these instructions are instantiated into multiple interconnected virtual control modules. These virtual control modules include a data acquisition and initialization module, a low-level physics field strong coupling calculation module, a fluid-thermal topology mapping and functional construction module, an asymmetric thermal boundary feedforward compensation module, and a nonlinear optimization and control module. The system includes a communication interface for high-frequency data exchange between the computer equipment and the sensors and low-level programmable logic controllers of the external seamless steel pipe piercing machine.

[0017] The data acquisition and initialization module is used to read the parameters of the seamless steel pipe piercing machine and construct the geometric envelope model and nonlinear constraint space of the main deformation zone; the underlying physical field strong coupling calculation module is used to extract the strain rate tensor and generate the second-order anisotropic thermal conductivity tensor, and perform partial differential numerical solution of the three-dimensional steady-state thermo-mechanical coupling heat conduction control equation to output the three-dimensional temperature scalar field and the three-dimensional thermal gradient vector field; the fluid-thermal topology mapping and functional construction module is used to construct the local divergence penalty function and the system-level optimization objective functional, and set physical constraints; the asymmetric thermal boundary feedforward compensation module is used to perform spatial coordinate mapping and trigger the multi-segment independent high-pressure cooling mechanism; the nonlinear optimization and control module is used to iteratively solve the constrained nonlinear optimization mathematical model, and convert the obtained global optimal process parameter combination into an electronic control signal and send it to the physical actuator.

[0018] This invention provides a method for optimizing process parameters of a seamless steel tube piercing mill based on temperature field prediction. It has the following beneficial effects: 1. This invention constructs a second-order anisotropic thermal conductivity tensor by extracting the strain rate tensor from the metal rheological velocity field, ensuring spatial alignment of the principal thermal conductivity axis with the principal deformation direction of the metal. The heat transfer calculation method based on the microscopic lattice distortion mechanism overcomes the limitation of traditional heat conduction models that define thermal conductivity as an isotropic scalar. Within the severely deformed perforated region, it accurately characterizes the differences in heat conduction in different spatial directions, improving the numerical solution accuracy of the temperature scalar field and the thermal gradient vector field, and providing a reliable data foundation for multiphysics optimization.

[0019] 2. This invention establishes a local divergence penalty function by calculating the inner product of the metal rheological velocity vector field and the thermal gradient vector field, and constructs a system-level optimization objective functional by performing volume integration on the main deformation region. The local divergence penalty function physically characterizes the degree of dispersion of the metal flow trajectory from the isothermal surface tangent, transforming the risk of microcrack initiation into a quantifiable topological coaxial index. By combining equipment and material physical constraints to perform nonlinear programming iteration, the globally optimal combination of process parameters is obtained, reducing thermal stress concentration from the macroscopic process setting stage and suppressing the initiation of defects in the billet.

[0020] 3. This invention monitors the local divergence penalty function and extracts the three-dimensional coordinates of abnormal nodes when the function value exceeds a critical threshold. It then uses a normal shortest distance mapping rule to inversely project the abnormal nodes onto the target tool surface. Based on the excessive deviation, it calculates the target convective heat transfer coefficient and triggers a multi-segment independent high-pressure cooling mechanism, forming an asymmetric local strong cooling boundary on the target tool surface. This directionally alters the heat dissipation rate in a specific region inside the tube blank, forcibly deflecting the direction of the thermal gradient vector field, thus promoting spatial geometric alignment between the thermal gradient vector field and the metal rheological vector field. Attached Figure Description

[0021] Figure 1 The flowchart shows the optimization method for process parameters of seamless steel pipe piercing mill based on temperature field prediction. Figure 2 This is a schematic diagram of the S500 step process of the present invention. Detailed Implementation

[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Example: Please see the appendix Figure 1 -Appendix Figure 2 This invention provides a method for optimizing process parameters of a seamless steel pipe piercing machine based on temperature field prediction, including a data acquisition and initialization module, a low-level physical field strong coupling calculation module, a flow-heat topology mapping and functional construction module, an asymmetric thermal boundary feedforward compensation module, and a nonlinear optimization and control module.

[0024] The data acquisition and initialization module establishes a communication connection with the seamless steel pipe piercing machine and reads the basic geometric parameters of the machine and the initial thermophysical properties of the billet material. The data acquisition and initialization module also receives the initial process parameters input by the operator to optimize the variable space.

[0025] The underlying physics field strongly coupled calculation module receives parameter data transmitted from the data acquisition and initialization module. This module has built-in plasticity and thermodynamics solution programs. It extracts the rheological velocity field of the tube blank within the main deformation zone and extracts the strain rate tensor based on this field. The module then updates the anisotropic thermal conductivity tensor of the tube blank material using the strain rate tensor. Finally, it feeds the anisotropic thermal conductivity tensor back into the heat conduction control equations to solve for the temperature field and thermal gradient vector field. The specific numerical discretization calculation process for the heat conduction control equations can be solved using the finite element method by those skilled in the art; this process is well-known in the field and will not be elaborated upon further in this specification.

[0026] The fluid-thermal topology mapping and functional construction module receives the rheological velocity field vector and thermal gradient vector field output from the underlying physics field strongly coupled calculation module. This module calculates the local divergence penalty function by performing a geometric space dot product operation on the rheological velocity field vector and the thermal gradient vector field. Finally, it performs a volume integral operation on the local divergence penalty function within the main deformation region spatial domain to construct the objective functional for nonlinear system optimization.

[0027] The asymmetric thermal boundary feedforward compensation module monitors the numerical variation range of the local divergence penalty function. When the local divergence penalty function exceeds a preset critical threshold, the module extracts the three-dimensional spatial coordinates of the abnormal metal streamline domain. The preset critical threshold is obtained by calibrating the high-temperature yield strength and microcrack initiation critical stress of the billet material. The module then projects the three-dimensional spatial coordinates back onto the outer surface of the piercing mill guide device or mandrel. Finally, the module generates a high-pressure cooling medium injection command, altering the local thermal convection heat transfer coefficient of the outer surface and reversing the spatial direction of the internal temperature gradient vector.

[0028] The nonlinear optimization and control module embeds a nonlinear programming algorithm solver. Under the boundary conditions of equipment torque constraints and material temperature constraints, the module iteratively solves the objective functional of the nonlinear system, outputting the optimal combination of piercing mill process parameters. The module then converts this optimal combination of process parameters into electrical control drive signals and sends them to the physical actuators of the seamless steel pipe piercing machine. These physical actuators specifically include the main drive shaft motor, the pressing hydraulic cylinder, and the guide plate spacing adjustment motor.

[0029] Based on the multi-field topology mapping optimization system detailed above, the multi-field topology mapping optimization system executes a perforation mill process parameter optimization method based on deformation-induced anisotropic thermal conductivity and flow-heat vector coaxiality mapping, specifically including the following operation steps: S100, the data acquisition and initialization module acquires the basic geometric parameters of the seamless steel pipe piercing machine, the initial thermophysical parameters of the billet material, and the preset process optimization variable space.

[0030] The S200 module, a strongly coupled underlying physics calculation module, selects a set of process parameters within the process optimization variable space. Based on the selected process parameters and fundamental geometric parameters, this module calculates the metal rheological velocity vector field of discrete nodes within the main deformation zone. Finally, it extracts the strain rate tensor from the metal rheological velocity vector field.

[0031] The S300 module, a strongly coupled underlying physics calculation module, generates a second-order anisotropic thermal conductivity tensor reflecting the lattice distortion effect based on the strain rate tensor. This module then substitutes the second-order anisotropic thermal conductivity tensor into the thermodynamically coupled partial differential equations to obtain the three-dimensional temperature field of the main deformation region. Finally, the module performs spatial differentiation on the three-dimensional temperature field to obtain the three-dimensional thermal gradient vector field.

[0032] The S400 module, which performs fluid-thermal topology mapping and functional construction, calculates the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field, and obtains a local divergence penalty function to characterize the degree of conflict between the metal flow trajectory and the isothermal surface normal.

[0033] The S500 asymmetric thermal boundary feedforward compensation module monitors the local divergence penalty function. When the local divergence penalty function exceeds a preset critical threshold in a specific spatial domain, the asymmetric thermal boundary feedforward compensation module extracts the corresponding tool surface position according to the normal shortest distance mapping rule and triggers a multi-segment independent high-pressure cooling mechanism to reverse the direction of the three-dimensional thermal gradient vector field.

[0034] The S600, S600, fluid thermal topology mapping and functional construction modules perform volume integral calculations on the local divergence penalty function in the main deformation zone to obtain the system-level optimization objective functional, and establish material peak temperature constraint conditions and equipment limit torque constraint conditions in combination with billet material properties and equipment bearing parameters.

[0035] The S700 nonlinear optimization and control module combines the equipment's ultimate torque constraint and the material's peak temperature constraint, employing a nonlinear programming algorithm to iteratively optimize the system-level objective functional. When the change in the system-level objective functional within adjacent iteration steps is less than the preset convergence tolerance, the nonlinear optimization and control module outputs the globally optimal combination of process parameters. This globally optimal combination of process parameters is then sent to the programmable logic controller (PLC) to adjust the main drive shaft motor speed and the stroke of the hydraulic cylinder.

[0036] When executing step S100 as described above, the multi-field topology mapping optimization system completes the definition of the optimization variable space and the initialization of the multi-field physical boundaries through the data acquisition and initialization module. The data acquisition and initialization module internally includes a basic parameter submodule, a boundary setting submodule, and an optimization space configuration submodule. The specific workflow of the data acquisition and initialization module is detailed as follows: S101, the basic parameters submodule reads the inherent geometric parameters of the seamless steel tube piercing mill and the initial size parameters of the tube blank. The inherent geometric parameters include the nominal diameter of the rolls, the two-dimensional generatrix contour coordinate matrix of the mandrel, and the reference distance to the mounting surface of the guide device. The specific physical structure of the guide device in the seamless steel tube piercing mill is a guide plate or guide disc. The initial size parameters include the outer diameter and length of the solid tube blank. The basic parameters submodule uses the rolling centerline of the seamless steel tube piercing mill as... Using the theoretical center point of the main deformation zone as the origin, the inherent geometric parameters and initial size parameters are mapped to a three-dimensional Cartesian coordinate system to construct the geometric envelope model of the main deformation zone. For the three-dimensional solid modeling and finite element meshing algorithm of the geometric envelope model, those skilled in the art can use the Delaunay triangulation algorithm. Mesh generation algorithms are well-known techniques in this field and will not be elaborated upon in this specification.

[0037] S102, The boundary setting submodule obtains the initial thermal property parameter set of the tube blank material. The initial thermal property parameter set includes the initial isotropic thermal conductivity of the tube blank material. Specific heat capacity under constant pressure and material density The boundary setting submodule sets the initial thermodynamic and kinematic boundary conditions for the main deformation zone. The initial thermodynamic boundary conditions include not only the final temperature of the high-frequency induction heating before the billet enters the deformation zone of the piercing mill, but also the natural convection heat transfer boundary between the billet surface and the external environment, and the contact heat conduction boundary between the billet surface and the contact surface between the billet surface and the rolls or mandrel of the seamless steel pipe piercing mill. The kinematic boundary conditions are defined as the initial axial propulsion velocity of the billet at the moment it is bitten by the rolls. The boundary setting submodule provides material constitutive equation parameters to the underlying physical field strong coupling calculation module. These material constitutive equation parameters characterize the plastic rheological resistance of metallic materials at high temperatures. For the fitting process of the Hansel-Spitter constitutive model parameters for high-temperature rheological resistance, those skilled in the art can obtain the parameters using constant strain rate hot compression tests. The fitting process of the constitutive model parameters is a well-known technique in the field and will not be elaborated further in this specification.

[0038] S103, the optimization space configuration submodule receives the set of process optimization variables input by the operator and constructs a nonlinear constraint space. (Process optimization variable set) It is expressed by the following formula: In the formula, This indicates the roll feed angle, which is used to control the helical advance pitch of the billet; This indicates the angular velocity of the roll rotation, which is used to determine the dynamic loading frequency of the metal's plastic deformation. The ellipticity of the deformation zone is represented by the geometric ratio between the roll spacing and the guide plate spacing. This represents the mandrel extension amount, which is used to adjust the initial intervention point coordinates for axial metal flow inside the tube blank. The optimization space configuration submodule is the set of process optimization variables. The independent variables in the equation are given upper and lower limits for the physically feasible region, and an inequality constraint set is constructed. In the formula This represents the lower bound of the physical feasible region of the variable set for process optimization. This represents the upper limit of the physically feasible region of the process optimization variable set. The physical calibration of the upper and lower limits of the physically feasible region is based on the mechanical adjustment limits of the piercing mill's main drive motor and the anti-slip biting mechanical conditions of the billet. (Process optimization variable set) A continuous multidimensional optimization search space is formed within the physically feasible domain, which provides a variable search benchmark for the iterative calculation of the nonlinear optimization and control module.

[0039] After completing the definition of the optimization variable space and the initialization of the multi-field physical boundaries, the multi-field topology mapping optimization system executes step S200. In step S200, the underlying physics field strongly coupled calculation module extracts the metal rheological velocity field and strain rate tensor driven by plastic kinematics based on the input process optimization variable set. The specific execution steps of the underlying physics field strongly coupled calculation module are detailed as follows: S201. The strongly coupled underlying physics calculation module selects the combination of process parameters for the current iteration within the multidimensional optimization search space constituted by the process optimization variable set. Based on the roll rotation angular velocity and roll feed angle in the process parameter combination, the module calculates the linear velocity vector of the roll contact surface. This linear velocity vector is then used as the frictional contact velocity boundary condition for the outer surface of the billet. The selected combination of process parameters, the frictional contact velocity boundary condition, and the kinematic boundary condition are then substituted into the three-dimensional plastic flow equation. Based on the principle of volume incompressibility of metallic materials during plastic deformation, the module calculates the rheological velocity vector field of discrete nodes in the main deformation zone in the Euler coordinate system. Metal rheological velocity vector field The mathematical expression is as follows: In the formula, Indicates metal node along The transverse flow velocity component in the axial direction; Indicates metal node along Radial flow velocity component in the axial direction; Indicates metal node along The axial extension velocity component in the axial direction; express Unit basis vectors along the axial direction; express Unit basis vectors along the axial direction; express The unit basis vector along the axis. For the nonlinear iterative solution algorithm of the velocity field in the process of metal plastic deformation, those skilled in the art can use the Newton-Raphson iteration method, which is a well-known technique in the field and will not be described in detail in the specification.

[0040] S202, the underlying physics field strongly coupled calculation module for the metal rheological velocity vector field Spatial differentiation is performed along a three-dimensional coordinate system. The underlying physics field strongly coupled calculation module extracts the strain rate tensor, which reflects the intensity of local deformation within the metal, through spatial differentiation. Strain rate tensor It is a second-order tensor, strain rate tensor The specific formulas for calculating the components are as follows: In the formula, the subscript With subscript The range of values ​​is , respectively corresponding to the three-dimensional Cartesian coordinate system axis, shaft and Axial direction; representing the strain rate tensor Shear strain rate components or normal strain rate components on the corresponding dimensional plane; and Corresponding to the rheological velocity vector field of metal On the coordinate axes with coordinate axes Velocity component in the direction; and Represents spatial position coordinates. Strain rate tensor. These constitute the core input variables for subsequent updates to the anisotropic thermal conductivity properties of the tube blank material.

[0041] After acquiring the metal rheological velocity vector field and the corresponding strain rate tensor, the multi-field topology mapping optimization system enters the dynamic thermal conductivity property update stage. The underlying physics field strongly coupled calculation module executes steps S301 to S303, generating a second-order anisotropic thermal conductivity tensor reflecting the lattice distortion effect within the tube blank material based on the strain rate tensor. The detailed workflow of the underlying physics field strongly coupled calculation module is as follows: S301. During the piercing process of seamless steel pipes, the billet metal undergoes severe plastic deformation. This plastic deformation causes the internal grains to elongate along the main deformation direction, forming a fibrous structure, accompanied by lattice distortion. Grain elongation and lattice distortion alter the scattering paths of phonons and electrons within the metal, resulting in a higher thermal conductivity parallel to the plastic flow direction compared to the initial state, while the thermal conductivity perpendicular to the plastic flow direction is lower. The underlying physics field strongly coupled calculation module constructs a formula for calculating the second-order anisotropic thermal conductivity tensor, controlled by the strain rate tensor. The specific formula for calculating the second-order anisotropic thermal conductivity tensor is as follows: In the formula, This represents the second-order anisotropic thermal conductivity tensor, which is used to replace the scalar thermal conductivity in traditional heat transfer calculations to characterize the differences in heat conduction in different directions in three-dimensional space. This represents the initial isotropic thermal conductivity of the tube blank material before plastic deformation occurs; Represents a second-order unit tensor; This represents the fourth-order deformable thermal coupling tensor; This represents the tensor double dot product operator; This represents the strain rate tensor extracted by the underlying physical field strong coupling calculation module in the preliminary step.

[0042] S302, Fourth-order deformable thermal coupling tensor This is used to quantitatively characterize the degree of reshaping effect of microscopic crystal structure distortion on macroscopic heat conduction paths. The underlying physics field strong coupling calculation module internally stores a pre-calibrated fourth-order deformable thermal coupling tensor. Fourth-order deformable thermally coupled tensor The values ​​of the internal tensor elements were determined by high-temperature deformation thermal conductivity calibration tests on the tube blank material. The specific procedures for these high-temperature deformation thermal conductivity calibration tests included: subjecting cylindrical specimens of the tube blank material to high-temperature hot compression under different strain rate gradients; independently measuring the thermal conductivity of the hot-compressed specimens in the direction parallel to the compression axis and perpendicular to the compression axis using a laser flash method; and calculating the fourth-order deformation thermal conductivity coupling tensor based on the mapping relationship between the thermal conductivity difference and the input strain rate using nonlinear mathematical fitting. The values ​​of the internal tensor elements.

[0043] S303, the strongly coupled underlying physics calculation module, establishes a dynamic heat conduction tensor at each discrete node in the main deformation region spatial domain using the second-order anisotropic heat conduction tensor calculation formula. To ensure the consistency of the physical coordinate system, the strongly coupled underlying physics calculation module calculates the strain rate tensor. The eigenvalues ​​and eigenvectors are obtained, and a three-dimensional orthogonal rotation transformation matrix is ​​constructed using the eigenvectors. The underlying physics field strongly coupled calculation module applies the three-dimensional orthogonal rotation transformation matrix to the second-order anisotropic thermal conductivity tensor, ensuring that the principal axes of the second-order anisotropic thermal conductivity tensor are aligned in real time with the principal deformation directions of the metal at the corresponding spatial locations. The underlying physics field strongly coupled calculation module completes the spatial distribution extraction of the second-order anisotropic thermal conductivity tensor on all discrete nodes, providing underlying physical property input data for solving the subsequent heat conduction control equations.

[0044] After establishing the second-order anisotropic thermal conductivity tensor for discrete nodes within the main deformation region, the multi-field topology mapping optimization system continues with the temperature field solution and thermal gradient extraction stages. The underlying physics field strongly coupled calculation module executes steps S304 to S306, utilizing dynamically evolving thermal conductivity properties and thermo-mechanical coupled partial differential equations to solve the three-dimensional nonlinear temperature field and further calculate and extract the three-dimensional thermal gradient vector field. The detailed workflow of the underlying physics field strongly coupled calculation module is as follows: S304, the underlying physics field strong coupling calculation module establishes the three-dimensional steady-state thermo-mechanical coupling heat conduction control equations for the main deformation region, including the second-order anisotropic thermal conductivity tensor. These three-dimensional steady-state thermo-mechanical coupling heat conduction control equations describe the balance between convective and conductive heat transfer inside the tube blank under severe deformation conditions. The specific mathematical expression of the three-dimensional steady-state thermo-mechanical coupling heat conduction control equations is as follows: In the formula, Indicates the density of the tube blank material; This indicates the constant-pressure specific heat capacity of the tube blank material; This represents the metal rheological velocity vector field extracted in the previous step; Indicates the temperature of the space node; Represents the Hamiltonian operator; This represents the second-order anisotropic thermal conductivity tensor calculated and obtained by the underlying physical field strongly coupled calculation module in the preceding steps; This represents the heat generation rate of the volumetric heat source generated by plastic deformation work. This represents the heat generation rate of the frictional volumetric heat source, which is converted from the frictional work at the interface between the tool and the tube blank.

[0045] S305, the underlying physical field strong coupling calculation module calculates the heat generation rate of the volumetric heat source during plastic deformation based on the mechanical state of the billet metal during the deformation process. The formula for calculating the heat generation rate of the volumetric heat source during plastic deformation is: In the formula, The Taylor-Quinni coefficient represents the proportion of solidification in which plastic work is converted into heat energy. The Cauchy stress tensor represents the internal stress of a metal. Represents the strain rate tensor; This represents the tensor double dot product operator. The underlying physics-driven strongly coupled calculation module calculates the frictional surface heat flux density based on the interfacial frictional shear stress and relative slip velocity between the billet surface and the seamless steel pipe piercing machine tool. This heat flux density is then equivalently transformed into the heat generation rate of the frictional volumetric heat source acting on the discrete nodes of the billet surface. The formula for calculating the heat flux density of the friction surface is as follows: In the formula, This represents the heat flux density of the friction surface; The distribution coefficient representing the conversion of frictional work into heat energy; This represents the interfacial frictional shear stress; This represents the relative sliding velocity between the metal surface of the tube blank and the surface of the tool in the seamless steel pipe piercing machine. The underlying physics field strongly coupled calculation module, combining the natural convection heat transfer boundary and contact heat conduction boundary initialized in the previous steps, as well as the previously calculated heat generation rates of the plastic deformation volume heat source and the friction volume heat source, performs numerical discretization to solve the three-dimensional steady-state thermo-mechanical coupling heat conduction control equations. For the nonlinear numerical solution process of the three-dimensional steady-state thermo-mechanical coupling heat conduction control equations, those skilled in the art can use the Galerkin weighted residual method, which is a well-known technique in the field and will not be elaborated further in this specification. The underlying physics field strongly coupled calculation module obtains the three-dimensional temperature scalar field of discrete nodes within the main deformation zone through nonlinear numerical solution. .

[0046] S306, the underlying physics field strongly coupled calculation module performs continuous partial derivative operations along the three spatial dimensions of the three-dimensional Cartesian coordinate system on the obtained three-dimensional temperature scalar field to generate a three-dimensional thermal gradient vector field. The formula for extracting the three-dimensional thermal gradient vector field is: In the formula, Represents the three-dimensional temperature scalar field along Temperature gradient component along the axial direction; Represents the three-dimensional temperature scalar field along Temperature gradient component along the axial direction; Represents the three-dimensional temperature scalar field along Temperature gradient component along the axial direction; , , Representing the three-dimensional Cartesian coordinate system axis, shaft and The unit basis vector along the axis. Since the second-order anisotropic thermal conductivity tensor participates in the underlying iterative calculation of the heat transfer physics process, the generated three-dimensional thermal gradient vector field inherently contains the dynamic feedback effect of the metal rheological velocity vector field in its physical mechanism. The three-dimensional thermal gradient vector field and the metal rheological velocity vector field are jointly output, constituting the core input data for the subsequent construction of the fluid-thermal topology mapping functional.

[0047] After acquiring the metal rheological velocity vector field and the three-dimensional thermal gradient vector field, the multi-field topology mapping optimization system enters the fluid-thermal topology mapping and functional construction stage. The fluid-thermal topology mapping and functional construction module executes steps S401 to S403, calculating the local divergence penalty function based on the physical characterization mechanism of tube blank defects. The detailed workflow of the fluid-thermal topology mapping and functional construction module is as follows: S401. During the piercing process of seamless steel pipes, the initiation of microcracks inside the billet is not only limited by the local absolute temperature value, but also depends on the physical matching relationship between the direction of metal plastic deformation flow and the spatial temperature distribution gradient. When the metal material traverses a spatial region with a steep temperature gradient at an extremely high rheological velocity, severe thermal stress coupled with shear stress is generated inside the billet. This thermal stress coupled with shear stress induces extremely uneven deformation within the material, ultimately leading to the initiation of microcracks at the center of the billet. The fluid-thermal topology mapping and functional construction module introduces a fluid-thermal vector coaxiality metric to evaluate the degree of geometrical conflict between the metal rheological velocity vector field and the three-dimensional thermal gradient vector field in three-dimensional space. Under the ideal thermo-mechanical co-deformation state, the metal flow trajectory should be parallel to the isothermal surface, that is, the metal rheological velocity vector should be perpendicular to the three-dimensional thermal gradient vector. When the thermo-mechanical co-deformation state occurs, the additional thermal strain brought about by crossing the temperature difference region is minimal.

[0048] The S402 module, which performs topological mapping and functional construction, extracts the metal rheological velocity vector field and three-dimensional thermal gradient vector field of discrete nodes in the main deformation region generated in the pre-processing steps by the strongly coupled physical field calculation module. It then constructs a local divergence penalty function at each discrete node within the spatial domain of the main deformation region. The local divergence penalty function is expressed as the absolute value of the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field. Physically, the local divergence penalty function is equivalent to the convective derivative of the spatial node temperature. It characterizes the degree of dispersion of the metal rheological velocity vector from the isothermal surface tangent, thereby quantitatively calculating the severity of the metal material's penetration of the isothermal surface. The mathematical expression of the local divergence penalty function is: In the formula, Represents the local divergence penalty function; Represents the rheological velocity vector field of metal; Represents a three-dimensional thermal gradient vector field; This represents the vector dot product operator; This represents the absolute value operator. The larger the value of the local divergence penalty function, the smaller the angle between the metal flow direction and the temperature gradient direction, the more significant the metal's behavior when traversing regions of severe temperature differences, and the higher the risk of microcracks forming inside the tube blank.

[0049] The S403 module, which performs a spatial expansion calculation of the local divergence penalty function in three-dimensional Cartesian coordinates, substitutes the components of the metal rheological velocity vector field and the components of the three-dimensional thermal gradient vector field into the expansion formula. The specific expression of the expansion formula for the local divergence penalty function is as follows: In the formula, , , Representing discrete nodes along Lateral flow velocity component along the axis, The radial flow velocity component along the axial direction and along The axial extension velocity component in the axial direction; , , Representing the three-dimensional temperature scalar field respectively axial direction, along axial direction and along The temperature gradient component along the axial direction. The fluid-thermal topology mapping and functional construction module traverses all discrete nodes within the spatial domain of the main deformation region, calculating the spatial numerical distribution matrix of the local divergence penalty function. The data structure of the spatial numerical distribution matrix contains the three-dimensional spatial coordinate indices of all discrete nodes and the calculated values ​​of the local divergence penalty function corresponding to each discrete node. The spatial numerical distribution matrix of the local divergence penalty function reflects the defect initiation tendency at different physical locations within the main deformation region and constitutes the underlying data foundation for subsequent system-level optimization objective functional integration calculations and the asymmetric thermal boundary feedforward compensation triggering mechanism.

[0050] After constructing the spatial numerical distribution matrix of the local divergence penalty function, the asymmetric thermal boundary feedforward compensation module of the multi-field topology mapping optimization system executes steps S501 to S503. The asymmetric thermal boundary feedforward compensation module monitors the numerical variation range of the local divergence penalty function and performs inverse spatial coordinate mapping when abnormal values ​​are detected. The detailed workflow of the asymmetric thermal boundary feedforward compensation module is as follows: The S501 asymmetric thermal boundary feedforward compensation module extracts the spatial numerical distribution matrix of the local divergence penalty function output by the fluid-thermal topology mapping and functional construction module, and obtains the high-temperature yield strength and microcrack initiation critical stress of the billet material. The asymmetric thermal boundary feedforward compensation module calculates a preset critical threshold based on the high-temperature yield strength and microcrack initiation critical stress. The mathematical formula for calculating the preset critical threshold is: In the formula, This indicates a preset critical threshold. This represents the critical stress for microcrack initiation in the tube blank material at the current average temperature; This indicates the high-temperature yield strength of the tube blank material at the current average temperature; The dimension conversion coefficient is represented by a value obtained by fitting the mapping relationship between the peak value of the local divergence penalty function and the critical stress ratio using historical piercing defect sample data from seamless steel pipe piercing machines. The dimension conversion coefficient is used to unify the physical units on both sides of the formula. The preset critical threshold is used to characterize the critical coaxial state of thermal flux to prevent microcrack initiation in the billet metal. The asymmetric thermal boundary feedforward compensation module traverses each discrete node in the spatial numerical distribution matrix, determining whether the calculated value of the local divergence penalty function corresponding to the discrete node is greater than the preset critical threshold.

[0051] S502. When the asymmetric thermal boundary feedforward compensation module detects that the calculated value of the local divergence penalty function is greater than a preset critical threshold, the asymmetric thermal boundary feedforward compensation module determines that the corresponding discrete node is in a coaxiality anomaly state. The asymmetric thermal boundary feedforward compensation module extracts all discrete nodes in the spatial domain that are in a coaxiality anomaly state, constructs the metal streamline spatial domain where the anomaly occurs, and records the three-dimensional spatial coordinate indices of all anomalous nodes in the metal streamline spatial domain where the anomaly occurs. The three-dimensional spatial coordinates of the anomalous nodes are defined as follows: In the formula, , , These represent the values ​​of the three coordinate axes of the abnormal node in the three-dimensional Cartesian coordinate system.

[0052] The S503 asymmetric thermal boundary feedforward compensation module initiates the shortest normal distance mapping rule based on the three-dimensional spatial coordinates of the abnormal node and calls the geometric envelope model of the main deformation zone constructed in the data acquisition initialization module. The geometric envelope model of the main deformation zone includes the three-dimensional spatial surface equations of the outer surfaces of the seamless steel pipe piercing mill guide device, the rolls, and the mandrel, all within a fixed global spatial reference frame. The asymmetric thermal boundary feedforward compensation module calculates the normal projection distance from the abnormal node to each three-dimensional spatial surface equation and compares all calculated normal projection distances, selecting the target tool surface with the smallest normal projection distance. The asymmetric thermal boundary feedforward compensation module projects the abnormal node along the normal direction to the target tool surface, obtaining the corresponding surface mapping coordinates of the target tool surface. For the calculation of the normal projection distance from a spatial point to a complex curved surface and the algorithm for finding the shortest distance, those skilled in the art can use the gradient descent method or analytical geometric differentiation algorithm. The algorithm for finding the shortest distance is a well-known technology in this field and will not be described in detail in this specification. The asymmetric thermal boundary feedforward compensation module completes the extraction of surface mapping coordinates, which constitute the physical positioning reference for subsequently triggering the multi-segment independent high-pressure cooling mechanism.

[0053] After obtaining the surface mapping coordinates corresponding to the target tool surface, the multi-field topology mapping optimization system continues to execute multi-segment independent high-pressure cooling control and thermal gradient vector torsion mechanism. The asymmetric thermal boundary feedforward compensation module executes steps S504 to S506, actively reshaping the spatial distribution of the temperature gradient inside the tube blank through asymmetric local strong cooling intervention. The specific workflow of the asymmetric thermal boundary feedforward compensation module is detailed below: S504, the asymmetric thermal boundary feedforward compensation module retrieves the hardware topology of the cooling system of the seamless steel pipe piercing mill. The cooling system hardware topology includes a multi-segment independent high-pressure cooling nozzle array embedded in the back arc surface of the guide device or the water trough inside the roll. The asymmetric thermal boundary feedforward compensation module performs spatial Euclidean distance matching between the surface mapping coordinates obtained in the previous step and the physical installation coordinates of the multi-segment independent high-pressure cooling nozzle array, and locks the independent high-pressure cooling nozzle whose spatial physical distance is closest to the surface mapping coordinates. This locked independent high-pressure cooling nozzle is then used as the target nozzle for performing cooling intervention.

[0054] The S505 asymmetric thermal boundary feedforward compensation module calculates the target convective heat transfer coefficient required by the target nozzle based on the degree of exceeding the limit of the local divergence penalty function. To change the normal direction of the isothermal surface inside the tube blank and make it tend to be parallel to the metal rheological velocity vector, the asymmetric thermal boundary feedforward compensation module constructs a nonlinear compensation equation for the convective heat transfer coefficient. The specific mathematical expression of the nonlinear compensation equation for the convective heat transfer coefficient is as follows: In the formula, This indicates the target convective heat transfer coefficient that the target execution nozzle needs to output; This represents the convective heat transfer coefficient of the target tool surface under the condition of not triggering abnormal compensation. The convective heat transfer coefficient of the base surface is obtained by querying the base pipe water pressure and standby flow parameters of the current cooling system in standby state through a preset fluid dynamics mapping table. This represents the heat transfer compensation gain coefficient. The specific value of the heat transfer compensation gain coefficient is obtained from the offline thermal-fluid coupling calibration experiment. The heat transfer compensation gain coefficient is used to characterize the convective heat transfer intensity required to eliminate the change in the unit local divergence penalty function. This represents the calculated value of the local divergence penalty function for discrete nodes in a state of coaxiality anomaly. This represents the preset critical threshold calculated and calibrated in the preliminary steps. The target convective heat transfer coefficient calculated by the nonlinear compensation equation is used to form a local asymmetric strong cooling boundary on the surface of the target tool. The asymmetric strong cooling boundary applies a directional heat flow suction effect to the interior of the tube blank through heat conduction. The heat flow suction effect changes the local heat dissipation rate of discrete nodes inside the tube blank, thereby deflecting the direction of the three-dimensional thermal gradient vector field inside the tube blank.

[0055] S506, the asymmetric thermal boundary feedforward compensation module converts the target convective heat transfer coefficient into the physical fluid control parameters of the target actuation nozzle. The asymmetric thermal boundary feedforward compensation module establishes a fluid dynamics mapping relationship between the cooling medium injection flow rate and the target convective heat transfer coefficient. The mathematical expression for this fluid dynamics mapping relationship is: In the formula, This indicates the required flow rate of cooling medium injected into the target nozzle; This represents the flow correction factor related to the hydraulic orifice diameter of the target nozzle; This indicates the constant-pressure specific heat capacity of the cooling medium; Indicates the density of the cooling medium; This indicates the jet heat transfer index. Flow correction factor. With jet heat transfer index The flow rate is determined by the factory parameters of the selected nozzle's hydrodynamic characteristics test. The asymmetric thermal boundary feedforward compensation module calculates the cooling medium injection flow rate and, combined with the pre-calibrated flow-valve opening characteristic curve of the electromagnetic proportional control valve, converts the cooling medium injection flow rate into an electronically controlled pulse signal containing duty cycle information. The asymmetric thermal boundary feedforward compensation module sends the electronically controlled pulse signal to the electromagnetic proportional control valve corresponding to the water supply pipeline of the target execution nozzle. The electromagnetic proportional control valve adjusts its internal valve opening according to the electronically controlled pulse signal, outputting the corresponding cooling medium injection flow rate. The cooling medium injection flow rate precisely covers the target tool surface, forming an asymmetric spatial thermal boundary. Without adjusting the global process parameters of the main drive motor, the asymmetric spatial thermal boundary reverses the direction of the three-dimensional thermal gradient vector field inside the tube blank, achieving spatial geometric alignment between the three-dimensional thermal gradient vector field and the metal rheological velocity vector field, eliminating internal abnormal coaxiality, and avoiding microcrack defects induced by local thermal stress.

[0056] After obtaining the spatial numerical distribution matrix of the local divergence penalty function, the multi-field topology mapping optimization system performs a global multiphysics constraint setting and system-level optimization objective functional construction phase. The fluid-thermal topology mapping and functional construction module executes steps S601 to S604, aggregating the discrete physical field data into system-level optimization indices and applying physical boundary constraints to ensure engineering safety. The detailed workflow of the fluid-thermal topology mapping and functional construction module is as follows: The S601 module, which performs flow-heat topology mapping and functional construction, extracts the three-dimensional geometric envelope volume of the main deformation region's spatial domain. Within this volume, it discretizes and sums the spatial numerical distribution matrix of the local divergence penalty function in conjunction with the finite element mesh volume micro-elements in the geometric envelope model. This approximates the three-dimensional spatial volume integral operation, constructing a system-level optimization objective functional with the process optimization variable set as its independent variables. The mathematical expression of the system-level optimization objective functional is: In the formula, This represents the system-level optimization objective functional; This represents the set of process optimization variables set by the data acquisition initialization module in the preceding steps; Represents the total volume of the three-dimensional geometric envelope of the spatial domain of the principal deformation region; This represents the metal rheological velocity vector field calculated under the drive of the process optimization variable set; This represents the three-dimensional thermal gradient vector field calculated under the drive of the process optimization variable set; This represents a volumetric element within the spatial domain of the main deformation region. The system-level optimization objective functional is used to quantify the total thermodynamic conflict dissipation energy accumulated due to the deviation of the metal rheological velocity direction from the temperature gradient direction throughout the entire perforation process.

[0057] S602. To ensure the safe and stable operation of the seamless steel pipe piercing machine and the metallurgical quality of the billet material, the thermal flux topology mapping and functional construction module, combined with material limit constraints, constructs a temperature boundary barrier. The thermal flux topology mapping and functional construction module iterates through the three-dimensional temperature scalar field output by the strongly coupled underlying physics calculation module to retrieve the local highest peak temperature and sets the material peak temperature constraint condition. The mathematical expression for the material peak temperature constraint condition is: In the formula, This represents the local maximum peak temperature extracted from the three-dimensional temperature scalar field. This indicates the permissible safe temperature limit of the tube blank material. The permissible safe temperature limit of the tube blank material is jointly calibrated by the solidus temperature and the overheat-sensitive temperature of the tube blank material. Specifically, the calibration rule is to extract the smaller value between the solidus temperature and the overheat-sensitive temperature of the tube blank material and subtract a preset safe temperature margin constant. The material peak temperature constraint condition is introduced into the optimization space as an inequality boundary condition. The material peak temperature constraint condition is used to prevent localized melting inside the tube blank or severe overheating oxidation defects on the tube blank surface caused by combinations of process parameters.

[0058] S603, the thermal flow topology mapping and functional construction module constructs the mechanical boundary barrier. Based on the Cauchy stress tensor space distribution within the main deformation zone and the frictional contact mechanical state of the seamless steel pipe piercing mill roll surface, this module integrally calculates the rolling torque required for the seamless steel pipe piercing mill to complete the piercing action. For the numerical mechanical calculation process of solving the rolling torque based on contact stress integration, those skilled in the art can use the contact element integration algorithm, which is a well-known technology in this field and will not be elaborated further in this specification. The thermal flow topology mapping and functional construction module reads the rated load parameters of the main drive motor of the seamless steel pipe piercing mill and constructs the equipment's ultimate torque constraint conditions. The mathematical expression of the equipment's ultimate torque constraint conditions is: In the formula, This represents the rolling torque calculated by the seamless steel pipe piercing mill under the current set of process optimization variables; This indicates the maximum allowable torque limit of the main drive motor of the seamless steel pipe piercing machine. The equipment's limit torque constraint is a parameter used to prevent extreme parameters from causing mechanical overload damage to the seamless steel pipe piercing machine due to nonlinear optimization and control module output.

[0059] The S604 module, which performs heat transfer topology mapping and functional construction, mathematically encapsulates the system-level optimization objective functional, material peak temperature constraints, and equipment limit torque constraints to construct a complete constrained nonlinear optimization mathematical model. The constrained nonlinear optimization mathematical model is specifically represented as follows: In the formula, This represents the globally optimal combination of process parameters that the system seeks to solve; This refers to finding the system-level optimization objective functional within the inequality constraint space. The parameter set positioning operation reaches the minimum value; This indicates that the constraint prefix is ​​applied. and These represent the lower and upper bounds of the physical feasible region within the inequality constraint set constructed in the preceding steps, respectively. The complete constrained nonlinear optimization mathematical model transforms the multiphysics topological coaxial matching mechanism into a standard engineering mathematical programming problem. After the constrained nonlinear optimization mathematical model is constructed, it is directly output to the nonlinear optimization and control module, forming the core benchmark for executing optimization iterations and controlling the underlying physical devices.

[0060] After completing the construction of the constrained nonlinear optimization mathematical model and the parallel monitoring of the asymmetric spatial thermal boundary feedforward compensation mechanism, the multi-field topology mapping optimization system enters the multi-field topology optimization solution and underlying equipment execution stage. The nonlinear optimization and control module executes steps S701 to S704, solving for the globally optimal combination of process parameters and distributing this combination to the physical execution mechanism of the seamless steel pipe piercing machine. The detailed workflow of the nonlinear optimization and control module is as follows: S701, the nonlinear optimization and control module calls the built-in sequential quadratic programming solver to process the constrained nonlinear optimization mathematical model established in the previous step. The sequential quadratic programming solver is used to handle mathematical programming problems that include nonlinear objective functionals and physical boundary constraints. The nonlinear optimization and control module sets the initial combination of process parameters for the optimization iteration. Maximum number of iterations limit and preset convergence tolerance The specific value of the preset convergence tolerance is determined by the minimum mechanical adjustment resolution limit calibration of the physical actuator of the seamless steel pipe piercing machine. The nonlinear optimization and control module uses the initial process parameter combination... Starting from the variable set for process optimization, an iterative search is carried out within the nonlinear constraint space. For the Hessian matrix update and quadratic programming subproblem solving algorithms involved in the sequential quadratic programming solution program, those skilled in the art can use the BFGS quasi-Newton method for calculation. The Hessian matrix update and quadratic programming subproblem solving algorithms are well-known technologies in this field and will not be elaborated further in this specification.

[0061] S702, the nonlinear optimization and control module iteratively updates the process optimization variable set in each iteration search. In conjunction with the underlying physics field strongly coupled calculation module and the fluid-thermal topology mapping and functional construction module, the current process optimization variable set is recalculated. The corresponding system-level optimization objective functional Local peak temperature and rolling calculation torque In the formula, the subscript... This indicates the current iteration step number. The nonlinear optimization and control module continuously checks the iteration termination conditions during the iterative search process. The specific iteration termination conditions include: the absolute value of the change in the system-level optimization objective functional within adjacent iteration steps is less than or equal to the preset convergence tolerance, or the current iteration step number... Reaching the maximum number of iterations limit The mathematical formula for determining the termination condition of an iteration is: In the formula, Indicates the first The system-level optimization objective function value calculated in the next iteration; Indicates the first The system-level optimization objective function value calculated in the next iteration; This indicates the preset convergence tolerance, which is used to control the computational precision of nonlinear optimization iteration to match the execution capabilities of the underlying hardware. Indicates the current iteration step; This indicates the maximum number of iteration steps. When the iteration termination condition is met, the nonlinear optimization and control module stops iterating, extracts the parameter set that minimizes the system-level optimization objective functional, and confirms the extracted parameter set as the globally optimal combination of process parameters. Global optimal combination of process parameters Includes optimal roll feed angle Optimal roll rotational angular velocity Optimal guide distance and optimal tip extension .

[0062] S703, the nonlinear optimization and control module combines globally optimal process parameters. The digital quantities are converted into analog electrical control signals and digital pulse signals that the seamless steel pipe piercing mill equipment can recognize and execute. The nonlinear optimization and control module, based on the mechanical transmission ratio equation of the main drive reduction gearbox of the seamless steel pipe piercing mill equipment, determines the optimal roll rotation angular velocity. This is converted to the target output frequency of the frequency converter. The nonlinear optimization and control module, based on the hydraulic cylinder stroke mapping linkage equation of the seamless steel pipe piercing machine, determines the optimal roll feed angle. This is converted into a simulated voltage representing the target displacement of the hydraulic cylinder used to adjust the feed angle. The nonlinear optimization and control module, based on the screw pitch transmission equation of the seamless steel pipe piercing machine, determines the optimal guide distance. With optimal tip extension This is converted into the position pulse count of the corresponding position servo driver. The mechanical transmission ratio equation of the main drive reduction gearbox, the hydraulic cylinder stroke mapping linkage equation, and the lead screw pitch transmission equation are inherent factory mechanical parameters of the seamless steel pipe piercing machine equipment.

[0063] The S704 nonlinear optimization and control module packages the generated target output frequency, target displacement analog voltage, and position pulse count into a set of equipment control instructions. This set is then sent to the programmable logic controller (PLC) of the seamless steel pipe piercing machine via the industrial Ethernet bus. The PLC parses the control instructions, drives the main drive variable frequency motor to operate at the target output frequency, drives the hydraulic servo proportional valve to control the feed angle hydraulic cylinder to reach the spatial position corresponding to the target displacement analog voltage, and drives the guide distance adjustment servo motor and the push rod thrust servo motor to move to the physical coordinates corresponding to the position pulse count. The PLC completes the closed-loop adjustment of the physical actuators, ensuring that the seamless steel pipe piercing machine operates according to the globally optimal combination of process parameters. Stable operation. Under stable operation, the rheological velocity vector field of the billet metal and the three-dimensional thermal gradient vector field maintain a high degree of spatial coaxial mapping relationship, which reduces the degree of thermal stress concentration at the underlying physical cooperative deformation mechanism level and inhibits the initiation of microcrack defects inside the billet metal.

[0064] Based on the detailed description of the multi-field topology mapping optimization system and its execution steps, the multi-field topology mapping optimization system is divided into several interrelated virtual control modules at the functional logic level. The multi-field topology mapping optimization system includes a data acquisition and initialization module, a low-level physics field strongly coupled calculation module, a fluid-thermal topology mapping and functional construction module, an asymmetric thermal boundary feedforward compensation module, and a nonlinear optimization and control module.

[0065] The data acquisition and initialization module obtains the basic geometric parameters of the seamless steel pipe piercing machine and the initial thermophysical parameters of the billet material. It maps the inherent geometric parameters and initial size parameters to a three-dimensional Cartesian coordinate system to construct the geometric envelope model and nonlinear constraint space of the main deformation zone.

[0066] The underlying physics-field strongly coupled calculation module extracts the strain rate tensor of the billet metal from the process optimization variable set, and generates a second-order anisotropic thermal conductivity tensor based on the strain rate tensor. The underlying physics-field strongly coupled calculation module substitutes the second-order anisotropic thermal conductivity tensor into the three-dimensional steady-state thermo-mechanical coupled heat conduction control equation for partial differential numerical solution, and outputs a three-dimensional temperature scalar field and a three-dimensional thermal gradient vector field.

[0067] The fluid-thermal topology mapping and functional construction module calculates the local divergence penalty function between the metal rheological velocity vector field and the three-dimensional thermal gradient vector field. This module performs a three-dimensional volume integral operation on the local divergence penalty function, constructing a system-level optimization objective functional with the process optimization variable set as independent variables. It also establishes peak temperature constraints and equipment limit torque constraints based on the billet material properties.

[0068] The asymmetric thermal boundary feedforward compensation module extracts the three-dimensional spatial coordinates of abnormal nodes when the local divergence penalty function exceeds a preset critical threshold, calculates the surface mapping coordinates, and triggers a multi-segment independent high-pressure cooling mechanism. The asymmetric thermal boundary feedforward compensation module also reverses the direction of the three-dimensional thermal gradient vector field inside the tube blank through an asymmetric strong cooling boundary.

[0069] The nonlinear optimization and control module uses a sequential quadratic programming solution program to iteratively solve the constrained nonlinear optimization mathematical model. This module obtains the globally optimal combination of process parameters, converts it into analog electrical control signals and digital pulse signals, and sends them to the main drive variable frequency motor, hydraulic servo proportional valve, and position servo drive of the seamless steel pipe piercing machine.

[0070] Multi-field topology mapping optimization systems rely on specific computer equipment. This equipment can be instantiated as an industrial control server, a graphics workstation, or a programmable logic controller (PLC) host computer. The computer equipment includes a processor, memory, communication interfaces, and a system bus. The processor, memory, and communication interfaces are electrically connected and communicate data via the system bus.

[0071] The processor constitutes the core computing and control center of a computer device. The processor reads and executes computer program instructions stored in memory to complete steps S101 to S704 as described above. At the underlying hardware implementation level, the processor encompasses various integrated circuit forms. Specifically, the processor can be a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), application-specific integrated circuit (ASIC), or field-programmable gate array (FPGA).

[0072] Memory provides storage space for data and programs. It stores the core processes of the operating system, the partial differential equation solution algorithms for various physics calculations, the nonlinear optimization program code, and the discrete node data of the three-dimensional spatial mesh generated during the punching machine's physics calculations. Specific hardware implementations of memory encompass both volatile and non-volatile storage devices, including random access memory (RAM), read-only memory (ROM), flash memory, solid-state drives (SSDs), and hard disk drives (HDDs).

[0073] The communication interface is used for high-frequency data exchange between the computer equipment and the temperature sensors, pressure sensors, and underlying programmable logic controllers of the external seamless steel pipe piercing machine. The physical hardware form of the communication interface includes an industrial Ethernet interface, an RS485 serial communication interface, or a controller area network (CLAN) bus interface. For the data transmission control protocol of the system bus and the physical layer data packet communication protocol of the communication interface, those skilled in the art can use the standard TCP / IP protocol or the Modbus RTU protocol. The configuration and handshake mechanism of the communication protocol are well-known technologies in the field and will not be described in detail in this specification.

[0074] As a carrier of software code, the medium that carries computer program instructions is a computer-readable storage medium. A set of computer program instructions is burned or written into the computer-readable storage medium. When the computer program instructions are called and executed by the processor of a computer device, the computer device completely executes the processing logic corresponding to steps S101 to S704 detailed above. The specific physical form of the computer-readable storage medium is a non-volatile storage medium, including but not limited to: portable universal serial bus flash drives, read-only optical discs, digital multifunction optical discs, portable hard drives, magnetic tape storage, and network-attached storage devices.

[0075] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction, characterized in that, Includes the following steps: S100: Obtain the basic geometric parameters of the seamless steel pipe piercing machine, the initial thermophysical parameters of the billet material, and the preset process optimization variable space; S200. Select a set of process parameters in the process optimization variable space, calculate the metal rheological velocity vector field of discrete nodes in the main deformation zone based on the selected process parameters and basic geometric parameters, and extract the strain rate tensor based on the metal rheological velocity vector field. S300. Generate a second-order anisotropic thermal conductivity tensor that reflects the lattice distortion effect based on the strain rate tensor. Substitute the second-order anisotropic thermal conductivity tensor into the thermo-mechanical coupling partial differential equation and solve to obtain the three-dimensional temperature scalar field of the main deformation region. Perform spatial differentiation on the three-dimensional temperature scalar field to obtain the three-dimensional thermal gradient vector field. S400. Calculate the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field to obtain the local divergence penalty function used to characterize the degree of conflict between the metal flow trajectory and the normal of the isothermal surface. S500 monitors the local divergence penalty function. When the local divergence penalty function is greater than the preset critical threshold, it extracts the three-dimensional spatial coordinates of the abnormal node, projects the abnormal node onto the surface of the target tool according to the normal shortest distance mapping rule to obtain the surface mapping coordinates, and triggers a multi-segment independent high-pressure cooling mechanism based on the surface mapping coordinates to reverse the direction of the three-dimensional thermal gradient vector field. S600. The volume integral of the local divergence penalty function in the main deformation zone is used to obtain the system-level optimization objective functional. The material peak temperature constraint and the equipment limit torque constraint are established by combining the billet material properties and the equipment bearing parameters. S700 combines the equipment's limit torque constraint and the material's peak temperature constraint, and uses a nonlinear programming algorithm to iteratively optimize the system-level optimization objective functional. When the absolute value of the change of the system-level optimization objective functional in adjacent iteration steps is less than or equal to the preset convergence tolerance, it outputs the globally optimal process parameter combination and sends the globally optimal process parameter combination to the programmable logic controller to adjust the speed of the main drive shaft motor and the stroke of the pressing hydraulic cylinder of the seamless steel pipe piercing machine.

2. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, Obtaining the basic geometric parameters of the seamless steel pipe piercing mill, the initial thermophysical properties of the billet material, and the preset process optimization variable space specifically includes the following steps: The nominal diameter of the rolls, the two-dimensional generatrix contour coordinate matrix of the mandrel, and the reference distance of the mounting surface of the guide device of the seamless steel pipe piercing mill are read. The rolling centerline of the seamless steel pipe piercing mill is used as the vertical axis of the coordinate system, and the theoretical center point of the main deformation zone is used as the origin of the coordinate system. The nominal diameter of the rolls, the two-dimensional generatrix contour coordinate matrix of the mandrel, and the reference distance of the mounting surface of the guide device are mapped to the three-dimensional Cartesian coordinate system to construct the geometric envelope model of the main deformation zone. Set the high-frequency induction heating end temperature of the main deformation zone, the natural convection heat transfer boundary, the contact heat conduction boundary, and the initial axial advance speed of the billet at the moment it is bitten by the rolls. Receive a set of process optimization variables including roll feed angle, roll rotation angular velocity, deformation zone ellipticity, and mandrel extension. Set a lower limit and an upper limit of the physical feasible region for each independent variable in the process optimization variable set, and construct an inequality constraint set.

3. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, The metal rheological velocity vector field of discrete nodes in the main deformation zone is calculated based on the selected process parameters and basic geometric parameters, and the strain rate tensor is extracted from the metal rheological velocity vector field. The specific steps include: The linear velocity vector of the roll contact surface is calculated based on the roll rotation angular velocity and roll feed angle in the process parameter combination, and the linear velocity vector of the roll contact surface is used as the boundary condition for the frictional contact velocity of the outer surface of the billet. Substituting the selected combination of process parameters, frictional contact velocity boundary conditions, and kinematic boundary conditions into the three-dimensional plastic flow equation, and based on the principle of volume incompressibility of metallic materials in the plastic deformation stage, the rheological velocity vector field of discrete nodes in the main deformation zone in the Euler coordinate system is calculated. By performing spatial differentiation of the metal rheological velocity vector field along a three-dimensional Cartesian coordinate system, the strain rate tensor reflecting the intensity of local deformation within the metal is extracted.

4. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, The process of generating a second-order anisotropic thermal conductivity tensor reflecting the lattice distortion effect based on the strain rate tensor specifically includes the following steps: Obtain the initial isotropic thermal conductivity of the tube blank material before plastic deformation, as well as the pre-calibrated fourth-order deformation thermal coupling tensor; The fourth-order deformation thermal conductivity coupling tensor and the strain rate tensor are subjected to tensor double dot product operation. The result of the tensor double dot product operation is added to the product of the initial isotropic thermal conductivity and the second-order unit tensor to construct the second-order anisotropic thermal conductivity tensor. The eigenvalues ​​and eigenvectors of the strain rate tensor are calculated. A three-dimensional orthogonal rotation transformation matrix is ​​constructed using the eigenvectors. The three-dimensional orthogonal rotation transformation matrix is ​​applied to the second-order anisotropic thermal conductivity tensor so that the principal axis of the second-order anisotropic thermal conductivity tensor is aligned in real time with the principal deformation direction of the metal at the corresponding spatial position.

5. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, The step of substituting the second-order anisotropic thermal conductivity tensor into the thermo-coupling partial differential equation to obtain the three-dimensional temperature scalar field of the main deformation region specifically includes the following steps: A three-dimensional steady-state thermo-mechanical coupling heat conduction control equation is established, which includes the second-order anisotropic thermal conductivity tensor, the density of the tube blank material, the constant-pressure specific heat capacity of the tube blank material, and the metal rheological velocity vector field. The heat generation rate of the volumetric heat source during plastic deformation is calculated based on the Cauchy stress tensor, strain rate tensor, and Taylor-Quinni coefficients inside the metal. The friction surface heat flux density is calculated based on the interfacial friction shear stress and relative slip velocity between the tube blank surface and the seamless steel pipe piercing machine tool. The friction surface heat flux density is then equivalently converted into the heat generation rate of the friction volume heat source acting on the discrete nodes of the tube blank surface. By combining the pre-defined natural convection heat transfer boundary and contact heat conduction boundary, the heat generation rate of the plastic deformation volume heat source and the heat generation rate of the friction volume heat source are substituted into the three-dimensional steady-state thermo-mechanical coupling heat conduction control equation for numerical discretization solution, and the three-dimensional temperature scalar field of discrete nodes in the main deformation zone is obtained.

6. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, The calculation of the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field yields a local divergence penalty function that characterizes the degree of conflict between the metal flow trajectory and the isothermal surface normal. This process specifically includes the following steps: At each discrete node in the spatial domain of the main deformation region, a local divergence penalty function is constructed using the absolute value of the inner product of the metal rheological velocity vector field and the three-dimensional thermal gradient vector field. The spatial dimension expansion calculation of the local divergence penalty function in the three-dimensional Cartesian coordinate system is performed, and the components of the metal rheological velocity vector field and the components of the three-dimensional thermal gradient vector field are substituted into the expansion formula. Traverse all discrete nodes within the spatial domain of the main deformation region, and calculate the spatial numerical distribution matrix of the local divergence penalty function. The data structure of the spatial numerical distribution matrix contains the three-dimensional spatial coordinate index of all discrete nodes and the calculated value of the local divergence penalty function corresponding to the discrete nodes.

7. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 6, characterized in that, After calculating and obtaining the spatial numerical distribution matrix of the local divergence penalty function, the method further includes the following steps: The high-temperature yield strength and critical stress for microcrack initiation of the tube blank material are obtained, and the preset critical threshold is calculated based on the high-temperature yield strength, critical stress for microcrack initiation and dimension conversion coefficient. When the calculated value of the local divergence penalty function is greater than the preset critical threshold, the three-dimensional spatial coordinates of the abnormal node are extracted, the normal projection distance from the abnormal node to each three-dimensional spatial surface equation is calculated, and the abnormal node is projected along the normal direction to the target tool surface with the smallest normal projection distance to obtain the surface mapping coordinates. The surface mapping coordinates are matched with the physical installation coordinates of the multi-segment independent high-pressure cooling nozzle array using spatial Euclidean distance to lock the target execution nozzle. The target convective heat transfer coefficient that the target execution nozzle needs to output is calculated based on the difference between the calculated value of the local divergence penalty function and the preset critical threshold. Establish a hydrodynamic mapping relationship between the cooling medium injection flow rate and the target convective heat transfer coefficient, calculate and obtain the cooling medium injection flow rate, and combine it with the pre-calibrated flow rate-valve core opening characteristic curve of the electromagnetic proportional control valve to convert the cooling medium injection flow rate into an electronic control pulse signal containing duty cycle information and send it out.

8. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, The system-level optimization objective functional is obtained by performing a volume integral calculation on the local divergence penalty function in the main deformation zone. Then, material peak temperature constraints and equipment limit torque constraints are established based on the billet material properties and equipment load-bearing parameters, specifically including the following steps: The three-dimensional geometric envelope volume of the main deformation zone spatial domain is extracted. Within the three-dimensional geometric envelope volume, the spatial numerical distribution matrix of the local divergence penalty function is discretized and summed in combination with the finite element mesh volume micro-elements in the geometric envelope model to approximately realize the three-dimensional spatial volume integral operation. A system-level optimization objective functional with the process optimization variable set as independent variable is constructed. The local highest peak temperature is retrieved by traversing the three-dimensional temperature scalar field. The smaller value between the solidus temperature and the overheat sensitive temperature of the tube blank material is extracted and subtracted from the preset safety temperature margin constant to set the material peak temperature constraint condition. Based on the spatial distribution of Cauchy stress tensor in the main deformation zone and the integral of the frictional contact mechanical state of the roll surface of the seamless steel pipe piercing mill, the rolling calculation torque is calculated, the rated load parameters of the main drive motor of the seamless steel pipe piercing mill are read, and the ultimate torque constraint conditions of the equipment are constructed.

9. The method for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction as described in claim 1, characterized in that, The globally optimal process parameter combination is sent to the programmable logic controller to adjust the speed of the main drive shaft motor and the stroke of the pressing hydraulic cylinder of the seamless steel pipe piercing machine. Specifically, this includes the following steps: Based on the mechanical transmission ratio equation of the main drive reducer of the seamless steel pipe piercing mill, the optimal roll rotation angular velocity in the global optimal process parameter combination is converted into the target output frequency of the frequency converter. Based on the hydraulic cylinder stroke mapping linkage equation of the seamless steel pipe piercing mill, the optimal roll feed angle in the global optimal process parameter combination is converted into the target displacement simulation voltage of the hydraulic cylinder that adjusts the feed angle. Based on the screw pitch transmission equation of the seamless steel pipe piercing machine, the optimal guide distance and optimal mandrel extension in the global optimal process parameter combination are converted into the position pulse number of the corresponding position servo driver. The target output frequency, target displacement analog voltage, and position pulse count are packaged into a set of equipment control instructions, which are then sent to the programmable logic controller of the seamless steel pipe piercing machine via an industrial Ethernet bus to drive the main drive variable frequency motor, hydraulic servo proportional valve, guide distance adjustment servo motor, and push rod thrust servo motor.

10. A system for optimizing process parameters of a seamless steel pipe piercing mill based on temperature field prediction, characterized in that, The seamless steel pipe piercing mill process parameter optimization system, which is used to execute the seamless steel pipe piercing mill process parameter optimization method based on temperature field prediction as described in any one of claims 1 to 9, includes: a computer device, which includes a processor, a memory, a communication interface, and a system bus, wherein the processor, memory, and communication interface are electrically connected and communicate with each other through the system bus; The processor is used to read and execute computer program instructions stored in memory to instantiate multiple interconnected virtual control modules; The multiple interconnected virtual control modules include: The data acquisition and initialization module is used to acquire the basic geometric parameters of the seamless steel pipe piercing machine and the initial thermophysical parameters of the billet material, and to construct the geometric envelope model and nonlinear constraint space of the main deformation zone; The underlying physical field strong coupling calculation module is used to extract the strain rate tensor of the tube blank metal and generate a second-order anisotropic thermal conductivity tensor. The second-order anisotropic thermal conductivity tensor is substituted into the three-dimensional steady-state thermo-mechanical coupling heat conduction control equation for partial differential numerical solution, and outputs a three-dimensional temperature scalar field and a three-dimensional thermal gradient vector field. The fluid-thermal topology mapping and functional construction module is used to calculate the local divergence penalty function between the metal rheological velocity vector field and the three-dimensional thermal gradient vector field. The local divergence penalty function is used to perform three-dimensional spatial volume integral operation to construct the system-level optimization objective functional. The material peak temperature constraint condition and the equipment limit torque constraint condition are established by combining the billet material properties and equipment load parameters, respectively. The asymmetric thermal boundary feedforward compensation module is used to extract the three-dimensional spatial coordinates of abnormal nodes when the local divergence penalty function is greater than the preset critical threshold, calculate the surface mapping coordinates according to the normal shortest distance mapping rule, and trigger a multi-segment independent high-pressure cooling mechanism. The nonlinear optimization and control module is used to iteratively solve the constrained nonlinear optimization mathematical model using a sequential quadratic programming solution program, obtain the globally optimal combination of process parameters, and convert the globally optimal combination of process parameters into analog electronic control signals and digital pulse signals and send them to the physical actuator. The communication interface is used for high-frequency data exchange between computer equipment and the sensors and underlying programmable logic controllers of external seamless steel pipe piercing machine equipment.

Images